Deeply Recycled Origins of Mantle Heterogeneity at the Northern East Pacific Rise

Deeply Recycled Origins of Mantle Heterogeneity at the Northern East Pacific Rise | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Physical Sciences - Article Deeply Recycled Origins of Mantle Heterogeneity at the Northern East Pacific Rise Molly Anderson, Gabriel Johnston, Michael Perfit, Alessandro Forte, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6173319/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Compositionally diverse lavas from the 8°20’ N seamount chain near the East Pacific Rise (EPR) record greater geochemical heterogeneity than observed along the entire northern EPR, including evidence for a 4-component mantle, and the first discovery of high-µ (high time-integrated 238 U/ 204 Pb) mantle source at the northern EPR. The amplitude of Pb, Sr, and Nd isotopic heterogeneity is on the scale of the Galapagos plume but in an area distal to a mantle plume. Our study combines isotopic and trace element compositions of near-EPR seamounts with recent mantle particle tracking models to identify deeply subducted origins for northern EPR mantle heterogeneity. We show that geochemical models (using He, Pb, Sr, Nd isotopes and trace elements) and geodynamic models (using time-dependent mantle flow simulations based on tomography) are self-consistent with tectonically recycled material feeding the northern EPR asthenosphere. These models provide a rare opportunity to assess geographic origins of mantle heterogeneity, tying the chemical and isotopic variability at the northern EPR to a paleo subduction zone at ~ 130 Ma, and reveal that one of the fastest spreading ridges on Earth samples deeply recycled components associated with whole mantle convection. Earth and environmental sciences/Solid Earth sciences/Geochemistry Earth and environmental sciences/Solid Earth sciences/Geodynamics Earth and environmental sciences/Solid Earth sciences/Petrology Mantle heterogeneity East Pacific Rise off-axis seamounts seismic tomography mantle convection particle tracking radiogenic isotopes trace elements Figures Figure 1 Figure 2 Figure 3 Figure 4 Discussion Mid-ocean ridge and ocean island lavas serve as recorders of the time-integrated geochemical evolution of the mantle, tracing the extent and variability of mantle heterogeneities and subduction inputs over time 1–4 . The full spectrum of mantle source compositions may be obscured, however, in lavas erupted on the spreading ridge-axis, because these melts are aggregated during ascent, and are further homogenized in shallow crustal melt lenses, especially at medium—fast spreading ridges 5–11 . Small-scale heterogeneity in the upper mantle is better sampled by examining lavas erupted off-axis 1,8,12–20 . Thus, off-axis seamounts near the EPR can be used to determine the origin and evolution of the Pacific mantle components over time, and to constrain the geodynamic controls of one of the fastest-spreading ridges on Earth. Although there is no canonical geophysical evidence for mantle plume-derived upwelling directly beneath the northern EPR 21 , some early studies associated widespread geochemical heterogeneity embedded in the sub-EPR mantle, and preferential formation of off-axis seamounts on the Pacific plate, with horizontal flow from plumes distal to the EPR 18,22 . Other studies have invoked long-term injections of recycled oceanic crust, sediments, lithosphere, or some combination of metasomatic recycled components into the mantle 1,3,15,16,18,20,23–28 . Regardless of the nature of the subducted component, geochemical models alone provide limited constraints on spatiotemporal origins of mantle heterogeneities. Recent time-dependent mantle flow predictions from seismic tomography reveal a broad, deep-seated, mantle-wide upwelling beneath the EPR 29,30 , which could provide a solution to the enigmatic origin of EPR mantle heterogeneity. The 8°20’ N seamount lavas exhibit extreme heterogeneity among major and trace element concentrations and radiogenic isotopes over short (sub-km) length scales (e.g., Figure S1), providing new constraints on northern EPR mantle heterogeneity 13 . The major and trace element concentrations, Pb, Sr, Nd and He isotope ratios are reported in Table S1. The compositional range of the full suite of seamount lavas spans the entire range – and in some cases extends beyond – previously defined values of the northern EPR MORB and its adjacent seamounts (Fig. 1 ). 8°20’ N seamount [La/Sm] N = 0.30 to 2.85 (northern EPR and adjacent seamounts = 0.22 to 3.4). In addition, 3 He/ 4 He ranges from 6.4–9.2 R A (where R A is the atmospheric ratio), and is broadly anti-correlated with 206 Pb/ 204 Pb, which ranges from 17.5–20.2 (Fig. 1 ). By contrast, the northern EPR and other adjacent seamounts have 3 He/ 4 He up to 8.7 R A and 206 Pb/ 204 Pb = 18.2–19.8 16 (Figure S2). Importantly, several samples from one seamount (Coral) define a new end-member component with HIMU characteristics, possessing high 206 Pb/ 204 Pb ~ 20.2 and 208 Pb/ 204 Pb ~ 40.2. This HIMU component is similar to that found in the French Polynesian islands 31–36 and at St. Helena 37 (Fig. 1 ), but has not previously been found at the EPR. Nature of the Multi-Component Mantle Feeding the Northern EPR Combined radiogenic He, Pb, Sr, and Nd isotopes form an important framework for discriminating between oceanic mantle sources. Because radiogenic isotopes do not fractionate during shallow-level or recent magma evolutionary processes, their ratios record mantle source trace element ratios (e.g., U/Pb, U + Th/He, Rb/Sr, and Sm/Nd) integrated over geologic time. High-density sampling of oceanic crust especially at off-axis seamounts reveals the EPR mantle is isotopically highly heterogeneous 13,16,18,19,47–50 . Some isotopically enriched but incompatible element depleted lavas erupt along the EPR crest 10,51,52 , although these compositions are mostly found off-axis. Some studies have invoked two mantle components in the northern sub-EPR mantle: an incompatible element depleted MORB mantle (DMM) and an incompatible element enriched source (EMM) 13,38,52–54 . Other studies have argued for at least three mantle components in the near-EPR mantle 15,23 . Our results show that the sub-EPR mantle in a region far from a mantle plume contains at least four end-member mantle components: 1) an incompatible element (Nb/La < 1) and isotopically depleted source (DMM; 206 Pb/ 204 Pb < 18.5); 2) an incompatible element depleted (Nb/La 20.1); 3) an incompatible element enriched (Nb/La > 1.5), isotopically enriched source (possibly EMII); and 4) an incompatible element enriched source with isotopic compositions similar to EMI (radiogenic Sr and Nd, and unradiogenic Pb). The discovery of HIMU lavas at the northern EPR is consistent with the composition of several off-axis seamounts near the EPR approaching HIMU characteristics 8,16 (Fig. 1 ). Traditionally mid-ocean ridge heterogeneities (all but DMM) are associated with enrichment of incompatible elements, but in this study depleted MORB (DMORB) alone span the entire Pb, Nd, and He isotopic range, indicating the depleted mantle is heterogeneous 17 . Commonly invoked models accounting for origins of mantle heterogeneity include injections of recycled oceanic crust and slab derived fluids/melts 15,20,35,36,55–58 , recycled sediments 15,35 , recycled metasomatized lithosphere 4,28,37 and recycled carbonates 24,25,59 . Further, the timing and location of subduction, and the origin and nature of recycled components in the EPR mantle have remained unclear due to the inherently heterogeneous nature of globally subducted lithosphere and sediments 60 . Active Upwelling Mantle Source for Heterogeneity Beneath the EPR For the first time, we investigate the origins of EPR mantle heterogeneities identified among the 8°20’ N seamount lavas (the HIMU end-member specifically) by using tomography-based time-reversed mantle convection models 61,62 . To track the time-reversed convective evolution of parcels of the mantle, seismic tomography images of present-day mantle structures are used as starting conditions 63,64 . For tomographically well-characterized regions like the Pacific, present-day tomographic models can be used to infer past mantle structures and mantle flow, through time-reversed reconstructions of the evolution of mantle convection using the Back-and-Forth Nudging (BFN) numerical method employed here 61,63 . Surface boundary conditions are provided by coupling the motions of rigid plates (using constraints from plate reconstructions) with the underlying mantle flow, and two depth-dependent viscosity profiles (V1 and V2) are considered 65,66 . The resulting reconstructed mantle flowlines (particle tracks) are shown relative to current plate orientations at the surface (Figure S3; Figure S4). The V1 viscosity mantle flow model 65 is favored here due to the thinner, lower-viscosity lithosphere and shallow asthenospheric layer, which make it ideal for environments like mid-ocean ridges characterized by thinner lithosphere. Our particle track simulation, based on the V1 mantle viscosity, is initialized at 8.409454 N and − 104.412 E, and starts 50 km below the present-day location of the 8°20’ N seamounts. The time-reversed particle trajectory reaches its maximum depth of ~ 2500 km (or ~ 387 km above the core-mantle boundary) within ~ 76 Ma (Figure S3). The particle then returns to the upper mantle around − 85° E, 27° N, placing a time constraint on subduction-driven entry into the convecting lower mantle at around 130 Ma. Taking into account slab descent times (between ~ 10 and ~ 20 Ma) across the upper mantle 68 , these results are consistent with independently constrained plate-history reconstructions that suggest the paleo-Mezcalera plate was subducting beneath the eastern edge of the paleo-Farallon plate forming the Caribbean-Guerrero arc 67 . Prior time-dependent tomographic mantle flow models have predicted deep-sourced active upwelling beneath the EPR 62 , the results from which are consistent with the relative stability/immobility of the EPR over the past 90 million years despite asymmetric far-field slab pull 29,30 , as is indicated by the new particle tracking models. Based on particle tracking simulations, we consider two possibilities for the origin of EPR mantle heterogeneity: (1) rapid, full-mantle convection has recently dredged up recycled material that has potentially been residing in the lower mantle as deep as 400 km above the core-mantle boundary, in which case these recycling models constrain the timescales for elemental transport from source to sink across the length of the mantle; or, (2) Early Cretaceous material from a paleo subduction zone was rapidly convected through the mantle to ~ 400 km above the core-mantle boundary, and returned to the upper mantle beneath the EPR over the past ~ 130 to ~ 150 million years, taking into account possible subduction transit times across the upper mantle 68 . At this time, we cannot rule out the possibility of recently-dredged, ancient, deep-mantle material feeding the EPR. However, the extent of mixing along the particle track inferred appears to be limited (supplementary discussion C). If the latter case is true, we can provide new lower-limit constraints on elemental cycling through the mantle. We combine radiogenic isotope evolution models with trace element mixing-melting models to demonstrate how recently subducted material could feasibly produce heterogeneity present in the mantle beneath the EPR today. Recent Recycling Origins for EPR Mantle Heterogeneity Assuming the EPR heterogeneity currently beneath the ridge has contributions from Mezcalera-Farallon subduction-related processes, the composition of the material subducting beneath the paleo-Farallon plate between ~ 150 and ~ 130 million years ago can be assessed by comparing radiogenic isotope ratios of the 8°20’ N seamount lavas with their trace element compositions. For the isotopic evolutionary model, we focus on the HIMU component because the 8°20’ N seamount HIMU closely resembles the end-member isotope ratios of St Helena and French Polynesia, whereas the alternative recycling-related components (EM) appear to be mixtures with DMM and therefore make it more difficult to determine a time frame of mixing that distinguishes a mantle end-member (Fig. 1 ). HIMU lavas are characterized by highly radiogenic 206 Pb/ 204 Pb (paired with unradiogenic 87 Sr/ 86 Sr), requiring high time-integrated U/Pb and Th/Pb but low Rb/Sr to generate extreme enrichment in 206 Pb and 208 Pb isotopes over time paired with only moderate enrichment in 87 Sr 1 . A widely favored model involves HIMU originating from ancient recycled oceanic crust which was hydrothermally altered before subduction and reintroduced to the mantle between 2.0–3.0 Ga 2,33,36 . This model is largely reliant on Pb extraction during subduction which would result in high U/Pb ratios in the altered crustal precursors integrated over billions of years during recycling. Sulfur isotopes further constrain recycling ages for some ocean island HIMU sources to the Archean 69–71 . Given the new mantle convection constraints presented here, this type of origin for the northern EPR HIMU would be possible if ancient reservoirs previously trapped in the deep mantle are being sampled recently by whole-mantle convection beneath the EPR. HIMU lavas also typically possess 3 He/ 4 He values lower than typical MORB (7–10 R A ), suggesting materials forming HIMU have high time-integrated U + Th/ 3 He relative to MORB 16,44,45 . However, on billion-year timescales, substantially more 4 He should have accumulated 35 . The decoupling of U/Pb from (U + Th)/He poses a problem for ancient altered subducted oceanic crustal origins of HIMU 36,72,73 . Several models potentially explain the relative uniformity of 3 He/ 4 He values among ocean island HIMU: (1) open-system behavior of helium due to its high diffusivity and equilibration with surrounding mantle 35 , (2) incorporation of relatively less-degassed mantle into OIB HIMU sources 32 , (3) significantly lower (U + Th)/He than anticipated from old subducted oceanic crust (i.e., the addition of carbonates to subducting crust), or (4) timescales of recycling to form some HIMU sources are significantly shorter than traditionally thought. The predicted mantle-flow velocity field (“mantle wind”) that is reflected in the particle trajectories requires the consideration of a relatively recent origin for EPR HIMU. If µ is sufficiently high upon subduction, HIMU 206 Pb/ 204 Pb ratios can be produced in < 300 million years 76,77 . For instance, the presence of carbonate veins in altered oceanic crust can significantly increase µ, or hydrothermal alteration on the seafloor increases the U/Pb of oceanic crust by preferentially leaching Pb 74,76 . To assess the possibility of a young origin of the 8°20’ N seamount HIMU, reverse Pb evolution models are presented (Fig. 2 ; Figure S5), starting with the 8°20’ N seamount HIMU present-day Pb isotope ratios, µ ( 238 U/ 204 Pb) of 21–60, and κ of 3–4. Although the EPR HIMU could be produced in ~ 240 Ma by extremely radiogenic Pb isotope ratios consistent with Pacific sediments, we favor a model where µ is closer to 60, consistent with old altered oceanic slabs containing carbonate veins 27,74 (Fig. 2 ). In this case, the northern EPR HIMU could have formed in the past 130 Ma, consistent with the V1 particle tracking model if the Pb isotopes are initially dominated by the presence of sediments (Fig. 2 ). Even if the mixture involved only small amounts of sediment (for example, 1–2 %) and 98–99% ceanic crust (Table S2), Pb isotopes would be dominated by sediments (e.g., GLOSS Pb ~ 20 ppm whereas MORB Pb ~ 0.48 ppm). By contrast, Sr and Nd would be impacted to a lesser extent (sediment Sr ~ 327 ppm whereas MORB Sr ~ 136 ppm; sediment Nd ~ 27 ppm whereas MORB Nd ~ 10 ppm) 60,78 , which would explain why the Sr and Nd isotope ratios of the HIMU lavas do not imply a sedimentary origin. The Pb, Sr, Nd, and He isotopes and trace element ratios of the seamount HIMU basalts are consistent with recycled oceanic slab origins, but the trace element concentrations are different from typical incompatible element enriched HIMU basalts such as those from St Helena (Fig. 3 ). For instance, although the measured U/Pb for the seamount HIMU lavas (µ ~ 21) is high compared to normal depleted mantle (µ ~ 5), the most incompatible trace elements are extremely depleted (Fig. 3 ). High Rb/Sr and U/Pb are expected to result from hydrothermal alteration of young crust near the ridge, although this alteration process alone is not sufficient to create HIMU without additional alteration upon subsequent subduction 20,79 . Subduction related dehydration processes typically are expected to remove Pb, Sr, Rb (Rb more than Sr), Ca, Ba, K, and light rare earth elements (e.g., Fig. 3 A), also resulting in high U/Pb and intermediate Rb/Sr 15 . Despite having trace element ratio characteristics of subducted oceanic crust, the seamount HIMU primitive mantle-normalized trace element compositions are more depleted than bulk recycled crust, unlike typical ocean island HIMU lavas like those found at St. Helena (Fig. 3 ). Variable degrees of melting alone cannot explain differences in trace element abundances between EPR seamount HIMU and ocean island HIMU (Fig. 3 ). Either the nature of the EPR HIMU source is different from (and more depleted than) plume-sourced HIMU, or melting, mixing, and metasomatic processes drastically differ between these settings. For instance, the difference may be attributed to the degree to which the peridotite dominated mantle is contaminated and metasomatized; the mid-ocean ridge mantle being peridotite-dominated, whereas ocean island plume mantle melts can be derived from blended peridotite-pyroxenite lithologies in which the enriched lithologies metasomatically induce trace element enrichment of surrounding peridotite. This ultimately masks the truly incompatible element depleted nature of the HIMU mantle source. If HIMU sources at ocean islands are typically dominated by incompatible element depleted oceanic crust, then we would still expect to see depletion in incompatible element concentrations among HIMU basalts worldwide 1 . Thus, we argue that the 8°20’ N seamount HIMU may be representative of primary, incompatible element depleted HIMU mantle sources whereas ocean island HIMU have been overprinted by incipient melts of enriched lithologies embedded in mantle peridotite. To produce the trace element patterns of the northern EPR HIMU consistent with a mixture of subduction-modified basalt and sediments (as inferred from the Pb evolution models), multi-stage melting may be required (Fig. 3 B, 4 ) 80,81 . In this scenario, a mixture of 1% GLOSS and 99% subduction-modified basalt is first melted beneath the EPR in the early stages of the seamount chain’s development (Stage 1). After the refractory residue from ridge melting has rafted off-axis, a second stage of melting off-axis melting responsible for forming the seamounts produces a 5–15% partial melt (Fig. 4 ) consistent with the HIMU trace element patterns observed at Coral seamount, such as extreme Pb, Ba, and Th depletion (Fig. 3 B). Alternatively, the highly depleted EPR HIMU trace element pattern could be formed by melting a mixture of 95% DMM with 5% recycled material (ie, the 1% GLOSS + 99% sediment mixture) (Figure S6). However, in this simple (batch melting) scenario, the off-axis HIMU melts are derived by ~ 20% partial melting, which may be unrealistic off-axis. Regardless of the detailed melting-mixture scenario, there are multiple ways to create incompatible element depleted HIMU sources given a starting mantle source that includes 1% GLOSS and 99% subduction-modified basalt inferred from Pb isotope models. Our study provides a new multidisciplinary solution to the long-standing problem concerning the origin of EPR mantle heterogeneity. Combined geochemical and geodynamic models imply relatively short timescales for recycling deep into the mantle followed by rapid upwelling today beneath the EPR. Lavas from the 8°20’ N seamount chain record greater geochemical heterogeneity than has been observed along axis, including evidence for a 4-component mantle, and the first discovery of a high-µ mantle source at the northern EPR. This HIMU component is uniquely incompatible element depleted compared with global ocean island HIMU, potentially representing the primary depleted nature of HIMU sources worldwide if they derive from recycled lithosphere. The combined models tie the origin of current sub-EPR mantle heterogeneity to a paleo subduction zone at ~ 130 to ~ 150 Ma. Geochemical, geodynamic, and plate reconstruction models are independently consistent with recently subducted, deeply recycled mantle heterogeneities feeding the northern EPR, which may mean that deep-seated mantle buoyancy drives the fastest spreading ridge on Earth. A full consideration of rapid convection timescales is needed in mass balance calculations for the evolution of the Pacific mantle and its role in global elemental cycling. Other localities along the EPR, Pacific-Antarctic Ridge, and Southwest Indian Ridge consist of highly heterogeneous lavas in locations distal to mantle plumes 54,83,84 . In such cases, heterogeneity has been attributed to melting of a marble-cake upper mantle style mixture, but new methods like those in this study have the potential to trace the surface origin of such mantle heterogeneities on a planetary scale. Declarations Acknowledgements: We thank the captains and crews of the R/V Atlantis (AT37-05 and AT42-06), the HOV Alvin and AUV Sentry teams for ensuring successful field operations. We are grateful to the other science team members of the OASIS I and II expeditions (Off-Axis Seamount Investigation at Siqueiros) including P. Gregg, D. Fornari, M. Smith, D. Geist, S. Shirey, C. Lundstrom, B. John, R. Parnell-Turner, H. Cabaniss, B. Boulahanis, E. McCully, C. Trim, V. Romano, Y.J. Tan, J. Albright, and Y. Zhan. We thank M. Lytle and K. Bermudez for invaluable analytical assistance, and greatly appreciate discussions with Veronique Le Roux, Matthew Leybourne, Stephen Elardo, Peter Barry, and Forrest Horton. This work was supported by the National Science Foundation (NSF OCE-MGG 1356610, NSF OCE-MGG 1356822, NSF OCE-MGG 1357150, NSF OCE-MGG 2001314), the Burnham Research Grant at Boise State University, Geological Society of America Lipman Award, and the University of Florida Graduate School Funding Fellowship. Helium isotope analyses at OSU were supported by NSF OCE-MGG1763255. Data Availability Statement: Data in support of this manuscript are available in supplemental data and the original data from the publication of 2016 OASIS I cruise samples are online at EarthChem https://doi.org/10.26022/IEDA/111616 Author Contributions: All coauthors assisted with manuscript revision and interpretation of results. MKA collected the samples, performed the laboratory experiments, created the mixing, isotope evolution, and melting models, conceptualized the final model, and wrote the initial manuscript draft with input from coauthors; GJ created and interpreted the geodynamic simulations, and conceptualized the final model; MRP conceived and supervised the study and collected the samples; AF created and interpreted the geodynamic simulations and supervised the study; GK performed the laboratory experiments and created mixing models; DWG performed the laboratory experiments and assisted with isotope evolution and melting models, JD assisted with model calculations, and provided key feedback for coordinating geochemical and geodynamic model results, VDW designed and supervised the study and collected the samples; PG designed the geodynamic models. References Zindler, A. & Hart, S. Chemical Geodynamics. Annual Reviews Earth Planetary Science 14 , 493–571 (1986). Hofmann, A. W. Mantle geochemistry: the message from oceanic volcanism. Nature 385 , 219–229 (1997). Stracke, A., Hofmann, A. W. & Hart, S. R. FOZO, HIMU, and the rest of the mantle zoo. Geochemistry, Geophysics, Geosystems 6 , 1–20 (2005). Sun, S. S. & McDonough, W. F. Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. Geol Soc Spec Publ 42 , 313–345 (1989). Hebert, L. B. & Montési, L. G. J. Generation of permeability barriers during melt extraction at mid-ocean ridges. Geochemistry, Geophysics, Geosystems 11 , 1–17 (2010). Katz, R. F. & Weatherley, S. M. Consequences of mantle heterogeneity for melt extraction at mid-ocean ridges. Earth Planet Sci Lett 335–336 , 226–237 (2012). Keller, T., Katz, R. F. & Hirschmann, M. M. Volatiles beneath mid-ocean ridges: Deep melting, channelised transport, focusing, and metasomatism. Earth Planet Sci Lett 464 , 55–68 (2017). Niu, Y., Regelous, M., Wendt, I. J., Batiza, R. & O’hara, M. J. Geochemistry of near-EPR seamounts: importance of source vs. process and the origin of enriched mantle component. Earth Planet Sci Lett 199 , 327–345 (2002). Phipps Morgan, J. Melt Migration Beneath Mid-Ocean Spreading Centers. Geophys Res Lett 14 , 1238–1241 (1987). Perfit, M. R. & Chadwick, W. W. Magmatism at mid-ocean ridges: Constraints from volcanological and geochemical investigations. in Geophysical Monograph Series vol. 106 59–115 (Blackwell Publishing Ltd, 1998). Sinton, J. M. & Detrick, R. S. Mid-ocean ridge magma chambers. J Geophys Res 97 , 197–216 (1992). Allan, J. F., Batiza, R. & Lonsdale, P. Petrology and Chemistry of Lavas from Seamounts Flanking the East Pacific Rise Axis, 21 deg N: Implications Concerning the Mantle Source Composition for Both Seamount and Adjacent EPR Lavas. American Geophysical Union Geophysical Monogram Series 255–282 (1987). Anderson, M. et al. Extreme Heterogeneity in Mid-Ocean Ridge Mantle Revealed in Lavas From the 8°20′N Near-Axis Seamount Chain. Geochemistry, Geophysics, Geosystems 22 , (2021). Batiza, R. & Vanko, D. Petrology of young Pacific seamounts. J Geophys Res 89 , 11235–11260 (1984). Dixon, J. E. et al. Light Stable Isotopic Compositions of Enriched Mantle Sources: Resolving the Dehydration Paradox. Geochemistry, Geophysics, Geosystems 18 , 3801–3839 (2017). Graham, D. W. et al. He, Pb, Sr and Nd isotope constraints on magma genesis and mantle heterogeneity beneath young Pacific seamounts. Contrib Mineral Petrol 99 , 446–463 (1988). Lambart, S. et al. Highly heterogeneous depleted mantle recorded in the lower oceanic crust. Nat Geosci 12 , 482–486 (2019). Niu, Y. & Batiza, R. Trace element evidence from seamounts for recycled oceanic crust in the Eastern Pacific mantle. Earth Planet Sci Lett 148 , 471–483 (1997). Zindler, A., Staudigel, H. & Batiza, R. Isotope and trace element geochemistry of young Pacific seamounts: implications for the scale of upper mantle heterogeneity. Earth Planet Sci Lett 70 , 175–195 (1984). Dixon, J. E., Leist, L., Langmuir, C. & Schilling, J.-G. Recycled dehydrated lithosphere observed in plume-influenced mid-ocean-ridge basalt. 420 , 385–389 (2002). Meyzen, C. M. et al. Isotopic portrayal of the Earth’s upper mantle flow field. Nature 447 , 1069–1074 (2007). Phipps Morgan, J., Morgan, W. J., Yu-Shen Zhang & Smith, W. H. F. Observational hints for a plume-fed, suboceanic asthenosphere and its role in mantle convection. J Geophys Res 100 , 12753–12767 (1995). Castillo, P. R. et al. Petrology and Sr, Nd, and Pb isotope geochemistry of mid-ocean ridge basalt glasses from the 11°45′N to 15°00′N segment of the East Pacific Rise. Geochemistry, Geophysics, Geosystems 1 , 1–40 (2000). Castillo, P. R., MacIsaac, C., Perry, S. & Veizer, J. Marine Carbonates in the Mantle Source of Oceanic Basalts: Pb Isotopic Constraints. Sci Rep 8 , 1–7 (2018). Castillo, P. R. The recycling of marine carbonates and sources of HIMU and FOZO ocean island basalts. Lithos 216–217 , 254–263 (2015). Donnelly, K. E., Goldstein, S. L., Langmuir, C. H. & Spiegelman, M. Origin of enriched ocean ridge basalts and implications for mantle dynamics. Earth Planet Sci Lett 226 , 347–366 (2004). Stracke, A., Bizimis, M. & Salters, V. J. M. Recycling oceanic crust: Quantitative constraints. Geochemistry, Geophysics, Geosystems 4 , 1–33 (2003). Weiss, Y., Class, C., Goldstein, S. L. & Hanyu, T. Key new pieces of the HIMU puzzle from olivines and diamond inclusions. Nature 537 , 666–670 (2016). Rowley, D. B. et al. Kinematics and dynamics of the East Pacific Rise linked to a stable, deep-mantle upwelling. Sci Adv 2 , 1–18 (2016). Rowley, D. B. & Forte, A. M. Kinematics of the East Pacific Rise Retrodicted From Pacific and Farallon/Nazca Subduction-Related Torques: Support for Significant Deep Mantle Buoyancy Controlling EPR Spreading. J Geophys Res Solid Earth 127 , (2022). Lassiter, J. C., Blichert-Toft, J., Hauri, E. H. & Barsczus, H. G. Isotope and trace element variations in lavas from Raivavae and Rapa, Cook-Austral islands: Constraints on the nature of HIMU- and EM-mantle and the origin of mid-plate volcanism in French Polynesia. Chem Geol 202 , 115–138 (2003). Parai, R., Mukhopadhyay, S. & Lassiter, J. C. New constraints on the HIMU mantle from neon and helium isotopic compositions of basalts from the Cook-Austral Islands. Earth Planet Sci Lett 277 , 253–261 (2009). Hauri, E. H. & Hart, S. R. Re-Os isotope systematics of HIMU and EMII oceanic island basalts from the south Pacific Ocean. Earth Planet Sci Lett 114 , 353–371 (1993). Hanyu, T. & Nakamura, E. Constraints on HIMU and EM by Sr and Nd isotopes re-examined. Earth Planets Space 52 , 61–70 (2000). Hanyu, T. & Kaneoka, I. The uniform and low 3He/4He ratios of HIMU basalts as evidence for their origin as recycled materials. Letters to Nature 390 , 273–276 (1997). Chauvel, C., Hofmann, A. W. & Vidal, P. HIMU-EM: The French Polynesian Connection. Earth Planet Sci Lett 110 , 99–119 (1992). Graham, D. W., Humphris, S. E., Jenkins, W. J. & Kurz, M. D. Helium isotope geochemistry of some volcanic rocks from Saint Helena. Earth Planet Sci Lett 110 , 121–131 (1992). Goss, A. R. et al. Geochemistry of lavas from the 2005-2006 eruption at the East Pacific Rise, 9°46’N-9°56N: Implications for ridge crest plumbing and decadal changes in magma chamber compositions. Geochemistry, Geophysics, Geosystems 11 , 1–35 (2010). Waters, C. L. et al. Sill to surface: Linking young off-axis volcanism with subsurface melt at the overlapping spreading center at 9°03’N East Pacific Rise. Earth Planet Sci Lett 369–370 , 59–70 (2013). Graham, D. W., Christie, D. M., Harpp, K. S. & Lupton, J. E. Mantle Plume Helium in Submarine Basalts from the Galapagos Platform. Science Reports 262 , 2023–2026 (1993). Harpp, K. S. & Weis, D. Insights Into the Origins and Compositions of Mantle Plumes: A Comparison of Galápagos and Hawai’i. Geochemistry, Geophysics, Geosystems 21 , (2020). Harpp, K. S. & White, W. M. Tracing a mantle plume: Isotopic and trace element variations of Galápagos seamounts. Geochem. Geophys. Geosyst 2 , (2001). Kurz, M. D. & Geist, D. Dynamics of the Galapagos hotspot from helium isotope geochemistry. Geochim Cosmochim Acta 63 , (1999). Kurz, M. D., Jenkins, W. J. & Hart, S. R. Helium isotopic systematics of oceanic islands and mantle heterogeneity. Nature 297 , 97–99 (1982). Graham, D. W. Noble Gas Isotope Geochemistry of Mid-Ocean Ridge and Ocean Island Basalts: Characterization of Mantle Source Reservoirs. in Noble Gases in Geochemistry and Cosmochemistry: Reviews in Mineralogy and Geochemistry 247–318 (2002). Graham, D. W., Michael, P. J. & Rubin, K. H. An investigation of mid-ocean ridge degassing using He, CO2, and δ13C variations during the 2005–06 eruption at 9°50′N on the East Pacific Rise. Earth Planet Sci Lett 504 , 84–93 (2018). Batiza, R., Niu, V. & Zayac, W. C. Chemistry of seamounts near the East Pacific Rise: Implications for the geometry of subaxial mantle flow. Geology 18 , 1122–1125 (1990). Fornari, D. J., Perfit, M. R., Allan, J. F. & Batiza, R. Small-scale heterogeneities in depleted mantle sources: near-ridge seamount lava geochemistry and implications for mid-ocean-ridge magmatic processes. Letters to Nature 331 , 511–513 (1988). Hall, L. S., Mahoney, J. J., Sinton, J. M. & Duncan, R. A. Spatial and temporal distribution of a C-like asthenospheric component in the Rano Rahi Seamount Field, East Pacific Rise, 15°-19°S. Geochemistry, Geophysics, Geosystems 7 , 1–27 (2006). Shimizu, K. et al. Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: Implications for the volatile content of the Pacific upper mantle. Geochim Cosmochim Acta 176 , 44–80 (2016). Perfit, M. R. et al. Small-scale spatial and temporal variations in mid-ocean ridge crest magmatic processes. Geology 22 , 375–379 (1994). Waters, C. L., Sims, K. W. W., Perfit, M. R., Blichert-Toft, J. & Blusztajn, J. Perspective on the genesis of E-MORB from chemical and isotopic heterogeneity at 9-10°n East Pacific Rise. Journal of Petrology 52 , 565–602 (2011). Niu, Y., Collerson, K. D., Batiza, R., Wendt, J. I. & Regelous, M. Origin of enriched-type mid-ocean ridge basalt at ridges far from mantle plumes: The East Pacific Rise at 11°20′N. J Geophys Res Solid Earth 104 , 7067–7087 (1999). Prinzhofer, A., Lewin, E. & Allegre, C. J. Stochastic melting of the marble cake mantle: evidence from local study of the East Pacific Rise at 12 50’ N. Earth Planet Sci Lett 92 , 189–206 (1989). Hanyu, T., Kaneoka, I. & Nagao, K. Noble gas study of HIMU and EM ocean island basalts in the Polynesian region. Geochim Cosmochim Acta 63 , 1181–1201 (1999). Kimura, J. I., Gill, J. B., Skora, S., van Keken, P. E. & Kawabata, H. Origin of geochemical mantle components: Role of subduction filter. Geochemistry, Geophysics, Geosystems 17 , 3289–3325 (2016). Shimoda, G. & Kogiso, T. Effect of Serpentinite Dehydration in Subducting Slabs on Isotopic Diversity in Recycled Oceanic Crust and Its Role in Isotopic Heterogeneity of the Mantle. Geochemistry, Geophysics, Geosystems 20 , 5449–5472 (2019). Chauvel, C. et al. Fani Maoré, a new “young HIMU” volcano with extreme geochemistry. Earth Planet Sci Lett 626 , (2024). Hoernle, K., Tilton, G., Le Bas, M. J., Duggen, S. & Garbe-Schönberg, D. Geochemistry of oceanic carbonatites compared with continental carbonatites: Mantle recycling of oceanic crustal carbonate. Contributions to Mineralogy and Petrology 142 , 520–542 (2002). Plank, T. & Langmuir, C. H. The chemical composition of subducting sediment and its consequences for the crust and mantle. Chem Geol 145 , 325–394 (1998). Glišović, P. & Forte, A. M. A new back-and-forth iterative method for time-reversed convection modeling: Implications for the Cenozoic evolution of 3-D structure and dynamics of the mantle. J Geophys Res Solid Earth 121 , 4067–4084 (2016). Glišović, P. & Forte, A. M. Reconstructing the Cenozoic evolution of the mantle: Implications for mantle plume dynamics under the Pacific and Indian plates. Earth Planet Sci Lett 390 , 146–156 (2014). Glišovic, P. & Forte, A. M. Two deep-mantle sources for Paleocene doming and volcanism in the North Atlantic. Proc Natl Acad Sci U S A 116 , 13227–13232 (2019). Johnston, G., Forte, A. & Glisovic, P. Time-Dependent Convective Modelling of DUPAL Signatures in the South Atlantic: Implications for LLSVP Source Region. in AGU Fall Meeting DI44B-07 (New Orleans, Louisiana, 2021). Mitrovica, J. X. & Forte, A. M. A new inference of mantle viscosity based upon joint inversion of convection and glacial isostatic adjustment data. Earth Planet Sci Lett 225 , 177–189 (2004). Forte, A. M. et al. Joint seismic-geodynamic-mineral physical modelling of African geodynamics: A reconciliation of deep-mantle convection with surface geophysical constraints. Earth Planet Sci Lett 295 , 329–341 (2010). Mann, P., Rogers, R. D. & Gahagan, L. Overview of plate tectonic history and its unresolved tectonic problems. in Central America: geology, resources, and hazards 205–241 (2007). Goes, S., Agrusta, R., van Hunen, J. & Garel, F. Subduction-transition zone interaction: A review. Geosphere 13 , 644–664 (2017). Cabral, R. A. et al. Anomalous sulphur isotopes in plume lavas reveal deep mantle storage of Archaean crust. Nature 496 , 490–493 (2013). Dottin, J. W., Labidi, J., Lekic, V., Jackson, M. G. & Farquhar, J. Sulfur isotope characterization of primordial and recycled sources feeding the Samoan mantle plume. Earth Planet Sci Lett 534 , (2020). Farquhar, J. & Jackson, M. Missing Archean sulfur returned from the mantle. Proc Natl Acad Sci U S A 113 , 12893–12895 (2016). Kogiso, T., Tatsumi, Y., Shimoda, G. & Barsczus, H. G. High μ (HIMU) ocean island basalts in southern Polynesia: New evidence for whole mantle scale recycling of subducted oceanic crust. J Geophys Res Solid Earth 102 , 8085–8103 (1997). Wei, X. et al. Co-Occurrence of HIMU and EM1 Components in a Single Magellan Seamount: Implications for the Formation of West Pacific Seamount Province. Journal of Petrology 63 , 1–19 (2022). Hauff, F., Hoernle, K. & Schmidt, A. Sr-Nd-Pb composition of Mesozoic Pacific oceanic crust (Site 1149 and 801, ODP Leg 185): Implications for alteration of ocean crust and the input into the Izu-Bonin-Mariana subduction system. Geochemistry, Geophysics, Geosystems 4 , (2003). Erhardt, A. M. et al. Assessing Sedimentary Detrital Pb Isotopes as a Dust Tracer in the Pacific Ocean. Paleoceanogr Paleoclimatol 36 , (2021). Hoernle, K. Geochemistry of Jurassic Oceanic Crust beneath Gran Canaria (Canary Islands): Implications for Crustal Recycling and Assimilation. JOURNAL OF PETROLOGY 39 , 859–880 (1998). Hanan, B. B. & Graham, D. W. Lead and Helium Isotope Evidence from Oceanic Basalts for a Common Deep Source of Mantle Plumes. Science (1979) 272 , 991–995 (1996). Salters, V. J. M. & Stracke, A. Composition of the depleted mantle. Geochemistry, Geophysics, Geosystems 5 , 1–27 (2004). Bach, W., Bernhard, P. E., Hart, S. R. & Blusztajn, J. S. Geochemistry of hydrothermally altered oceanic crust: DSDP/ODP Hole 504B-Implications for seawater-crust exchange budgets and Sr-and Pb-isotopic evolution of the mantle. Geochemistry, Geophysics, Geosystems 4 , 1–29 (2003). Graham, D. W. & Michael, P. J. Predominantly recycled carbon in Earth’s upper mantle revealed by He-CO2-Ba systematics in ultradepleted ocean ridge basalts. Earth Planet Sci Lett 554 , 116646 (2021). Leybourne, M. I. & Van Wagoner, N. A. Heck and Heckle seamounts, northeast Pacific Ocean: high extrusion rates of primitive and highly depleted mid-ocean ridge basalt on off- ridge seamounts. J Geophys Res 96 , (1991). White, W. M. Trace Elements. in Geochemistry 258–312 (2013). Georgen, J. E., Kurz, M. D., Dick, H. J. B. & Lin, J. Low 3He/4He ratios in basalt glasses from the western Southwest Indian Ridge (10°-24°E). Earth Planet Sci Lett 206 , 509–528 (2003). Hamelin, C. et al. Geochemical portray of the Pacific Ridge: New isotopic data and statistical techniques. Earth Planet Sci Lett 302 , 154–162 (2011). Anderson, M. K. et al. Constraints on near-ridge magmatism using 40Ar/39Ar geochronology of enriched MORB from the 8°20’ N seamount chain. Earth Planet Sci Lett 618 , 118278 (2023). Romano, V. et al. The formation of the 8˚20’ N seamount chain, east pacific rise. Marine Geophysical Research 43 , (2022). Kamenov, G. D., Perfit, M. R., Mueller, P. A. & Jonasson, I. R. Controls on magmatism in an island arc environment: Study of lavas and sub-arc xenoliths from the Tabar-Lihir-Tanga-Feni island chain, Papua New Guinea. Contributions to Mineralogy and Petrology 155 , 635–656 (2008). Graham, D. W., Hanan, B. B., Hémond, C., Blichert-Toft, J. & Albarède, F. Helium isotopic textures in Earth’s upper mantle. Geochemistry, Geophysics, Geosystems 15 , 2048–2074 (2014). Simmons, N. A., Forte, A. M., Boschi, L. & Grand, S. P. GyPSuM: A joint tomographic model of mantle density and seismic wave speeds. J Geophys Res Solid Earth 115 , (2010). Glišović, P., Forte, A. M. & Moucha, R. Time-dependent convection models of mantle thermal structure constrained by seismic tomography and geodynamics: Implications for mantle plume dynamics and CMB heat flux. Geophys J Int 190 , 785–815 (2012). Glišović, P. & Forte, A. M. Importance of initial buoyancy field on evolution of mantle thermal structure: Implications of surface boundary conditions. Geoscience Frontiers 6 , 3–22 (2015). Methods 8°20’ N Seamount Sample Collection and Preparation The 8°20’ N seamount chain is an ~170 km long chain of volcanic ridges and seamounts oriented perpendicular to the EPR and north of the Siqueiros fracture zone (Figure S1). Two research expeditions on RV Atlantis (AT37-05 and AT42-06) in 2016 and 2018 sampled, mapped, and collected gravity and magnetic data across the chain. Two of the most geochemically heterogeneous seamounts, Oscar and Coral, were selected for highest-density sampling in 2018. Off-axis eruptions up to 90 km from the ridge axis are attributed to melt channelization from zones of weakness in the lithosphere related to the tectonic evolution of the nearby Siqueiros transform. The 8°20’ N seamounts form an E-W trending chain of volcanoes and constructional volcanic ridges extending ~170 km perpendicular from the northern East Pacific Rise. The chain is northwest of the Siqueiros transform and directly north of the Siqueiros fracture zone off-axis. Two research expeditions (AT37-05 and AT42-06) visited the 8°20’ N seamounts in 2016 and 2018 respectively, to conduct high spatial-resolution sampling using HOV Alvin (sample names starting with ALV) and dredges (sample names starting with OS). Although the 2016 cruise sampled the full length of the chain, the 2018 cruise focused on more detailed sampling along Oscar and Coral seamounts. Combined, the research expeditions collected 305 basalt samples with glassy rinds, all of which have been analyzed for major elements, more than half for trace elements (184 samples), 72 for Pb, Sr, and Nd isotopes, and 36 for He isotopes. The major and trace element concentrations and Pb, Sr, and Nd isotopes from the 2016 expedition are used to test models for melting heterogeneous mantle 13 . New 40 Ar/ 39 Ar geochronological and geophysical observations (magnetic and gravity data) from the 8°20’ N seamounts tie the seamount magmatism to the nearby Siqueiros and its complex tectonic evolution over the past ~3.8 myrs 85,86 . This paper is the first publication to include any He isotope data from the seamounts and new samples collected in 2018. The basalt samples are mostly clean aphyric basalts with glassy rinds useful for preserving the liquid composition of magma at the time of eruption. Alteration and phenocryst-free glasses were ultrasonically cleaned in deionized water. Trace Element Analytical Methods Trace element concentrations were determined on 148 samples using a Laser Ablation (UP213 Nd-YAG New Wave Research laser) ThermoElectron X-Series II Quadrupole Inductively Coupled Plasma Mass Spectrometer (LA-ICP-MS) at Boise State University 13 . The remaining 35 samples were measured on solutions using an Element II High-Resolution Inductively Coupled Plasma Mass Spectrometer at University of Florida. About 50 mg of clean alteration-free glass for each sample was digested in sealed Teflon vials at 100° C in HF-HNO 3 mixture for 24 hours. The solutions were evaporate dried, then acidified with 6N HCl while sealed at 100° C overnight. After evaporate drying again, the samples were diluted with 4.5 mL of 8 ppb Re-Rh spiked 5% HNO 3 and sealed at 100° C overnight. The samples were analyzed using Re and Rh as internal standards, and results were corrected using USGS standards AGV-1 and ENDV (resulting in a correction factor of 1.08 and 0.84 for Lu and Pb respectively). Pb, Nd, and Sr isotope Analytical Methods Radiogenic isotopes were collected on 72 basalts from the 8°20’ N seamounts at the University of Florida following previously established dissolution and chromatographic procedures 13,38 . Pb, Sr, and Nd isotopes were measured using a Nu-Plasma HR multicollector (MC) ICP-MS following methods described in refs 38,87 . Standards NBS-981, NBS-987, and JNdi-1 were run every 5-6 samples for Pb, Sr, and Nd respectively, and averages are reported in Table S3. Helium Analytical Methods Helium concentrations and 3 He/ 4 He ratios in basalt glass were determined at Oregon State University using in-vacuo crushing experiments followed by noble gas mass spectrometry 46,88 . Several clean, alteration-free chips of glass were loaded into stainless steel crushers where they were crushed 75 times while connected to the vacuum line. Released CO 2 and H 2 O were condensed together in a U-trap held at 77 K (liquid N 2 temperature) and non-condensable reactive gases were removed using SAES Zr-Al getters. Noble gases were cryogenically separated using a charcoal trap at 10 K. The trap was heated to 45 K, retaining neon and heavier noble gases and releasing helium directly into the Nu Noblesse Mass Spectrometer for isotope ratio and peak height measurements. Line blanks performed prior to sample analyses were typically < 5x10 -11 cm 3 STP 4 He. SMB-GLOSS Isotopic Mixing Calculations Based on trace element estimates for subduction-modified basalts (SMB 27 ) and isotope ratios for altered oceanic crust (AOC 74 ), and global subducting sediment trace elements and isotope ratios (GLOSS 60 ), we approximate the full range of possible Pb, Sr, and Nd isotope ratios for a subducted mixture of 99% SMB + 1% GLOSS required for our isotopic evolution model. While it is challenging to perfectly reconstruct a mixture consistent across all isotopic systems, we demonstrate in Table S2 a range of scenarios capable of producing a protolith required to form the HIMU within 130 Ma. In general, assuming there is very little sediment addition (1-2%) to subducting oceanic crust, Pb would be most significantly leveraged by sediments, while Sr and Nd would be less so. In the simplest scenario, we assume 1% average GLOSS Pb, Sr, and Nd are mixed with 99% of SMB, which successfully produces 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 208 Pb/ 204 Pb consistent with ratios our reverse Pb model predicts subducted ~130 Ma. However, in this simplified scenario using the GLOSS average, 87 Sr/ 86 Sr and 143 Nd/ 144 Nd of this mixture are too low to be reasonable protoliths. However, as ref 60 demonstrated, there are extremely large ranges of Pb, Sr, and Nd concentrations and isotope ratios even between trenches around the globe. To form our HIMU Sr and Nd ratios in 130 Ma, we instead require subducting sediment compositions closer to the lower end of GLOSS (Table S2), such as pelagic sediments more consistent with a mid-ocean arc distal to a continental sediment source. Particle Tracking Methods We reconstruct the past 3D structure of Earth’s mantle using numerical models of backward convective flow, with present-day mantle tomography as the initial condition 61,62 . By reversing time in the convection equations, we can trace the evolution of thermal anomalies and track how mantle parcels move from the present day back to earlier geologic epochs. This time-reversal approach employs convection models in which temperature and velocity fields are iteratively adjusted so that, when integrated forward again, they remain consistent with present-day tomography and (optionally) geologic constraints on plate motions. These methods provide a self-consistent framework that bridges global seismic tomography, mineral-physical constraints, geodynamic data, and plate reconstructions 63 . Model Setup : Our primary input is a 3D temperature anomaly field derived from the GyPSuM global tomography model 89 . Because tomography constrains lateral variations but not the horizontally averaged temperature (geotherm), we superimpose these anomalies on an adiabatic background (1600 K at the surface to ~2456 K at the core–mantle boundary: CMB). Key reference characteristics of the mantle (viscosity, density, thermal conductivity, etc.) are taken from refs 61,90 . Mantle internal heat sources (radioactive + secular cooling) sum to ~24 TW. We employ two depth-dependent viscosity profiles that are constrained by geodynamic observables. V1 model 65 has a lower viscosity and a thinner lithosphere, well-suited for modelling convective flow below oceanic plates and mid-ocean ridge regimes. V2 model 66 has an overall higher-viscosity mantle, thicker lithosphere, a low-viscosity layer at ~220 km, and a stiffer lower mantle. These viscosity models have been validated against mineral-physics constraints and surface geodynamic data 63,65,66 . Varying viscosity between V1 and V2 mainly shifts the timescales of mantle transport but does not substantially alter the large-scale geometry of the flow paths. Forward and Backward Convection : In forward mode, we implemented a pseudo-spectral solution of the conservation of mass, momentum, and energy for a Newtonian viscous fluid in a compressible, self-gravitating mantle 90 . Time-reversed flow calculations are numerically challenging due to the irreversibility of diffusion. We adopt the modified quasi-reversibility (QRV) technique 62 , which introduces a small biharmonic diffusion term to stabilize backward integration. We vary the regularization parameter with time, following the scheme in ref 61 , ensuring minimal unphysical diffusion while preventing exponential blow-up. To extend the accuracy of our time-reversed reconstructions, we couple the QRV method with the ‘Back-and-Forth Nudging’ (BFN) technique 61 . This iterative method alternates between forward and backward integrations of the mantle convection model, spanning 2.5 Myr time windows, and iteratively updates the temperature field at each time step to minimize cumulative errors associated with thermal diffusion. The BFN method preserves the resolution of mantle heterogeneity given by the present-day tomography model, and it allows for more precise reconstructions of mantle dynamics extending into the geologic past, often to 70 Ma and as far back as 250 Ma (see supplementary discussion A-C). Boundary Conditions, Model Resolution, and Plate Reconstructions : A plate-like mechanical boundary is applied at Earth’s surface, so the mantle’s buoyancy forces drive the plates rather than the plates imposing motion on the underlying mantle. Plate geometries over the Cenozoic are based on 5 Myr interval reconstructions in the Indo-Atlantic hotspot reference frame 29,30 . The models use: spherical harmonic expansions to degree 170 for horizontal resolution, yielding a horizontal resolution of ~64 km at the CMB and ~117 km near the surface; Chebyshev expansions up to order 129 radially, yielding a vertical resolution of 0.4 km near boundary layers to ~35 km in mid-mantle). Particle Tracking : Once the flow fields (“mantle wind”) are established, we compute particle flow lines via an eighth-order Dormand–Prince Runge–Kutta scheme that was implemented in ref 63 . These particle tracking calculations benefit from the enhanced precision afforded by the spectral method, which yields accurate calculations of the flow velocity at any point in the mantle domain. We note that deep mantle flow is relatively stable, preserving broad features (notably the “superplume” upwellings) for tens to hundreds of millions of years 90,91 . For times before 70 Ma, we often approximate deep flow as quasi-steady because the high viscosity in the lower mantle stabilizes long-wavelength downwelling and upwelling structures 90,91 . We compare these results to time-reversed mantle reconstructions that extend back to 250 Ma, which are subject to significant modelling uncertainties. A discussion of these associated uncertainties is presented in the supplementary discussion A. Application to the East Pacific Rise : We applied this approach to a mantle parcel at 8.409454 N, −104.412 E, 50 km beneath the 8°20’ N seamount chain near the East Pacific Rise (EPR). Under the V1 (low viscosity) profile, this particle descends to ~2500 km depth (~387 km above the CMB) in ~75 Myr (Figure S3). It then returns to the upper mantle at ~130 Ma, near −96° E, 15° N. These tracks align with independent reconstructions showing the paleo-Mezcalera plate subducting beneath the paleo-Farallon plate 67 , forming the Caribbean–Guerrero arc. Using the V2 (higher viscosity) profile almost doubles the transport time (~240 Ma to reach the same depth). However, both profiles converge on the same subduction-zone source region, demonstrating that while viscosity changes the timescale, it does not alter the essential pathway. Prior time-dependent studies 29,62 similarly suggest a long-lived upwelling beneath the southern half of the EPR, consistent with our backward tracking results. Hence, two scenarios could explain the recycled material feeding the EPR: (1) Rapid dredging of ancient deep-lower-mantle heterogeneities (~387 km above the CMB); (2) Oceanic crust subducted during the Early Cretaceous that transited the entire mantle over ~130 Myr. In either case, these particle tracks impose new lower-limit constraints on elemental cycling in Earth’s mantle. Combined with radiogenic isotope and trace element modeling, they support a role for subduction-driven heterogeneities in generating the notable geochemical signals at 8°20’ N seamounts. Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryDataTable1.xlsx Supplementary Data Table 1 SupplementaryDataTable2Mixingcalculations.xlsx Supplementary Data Table 2 SupplementaryMaterialfigure1to8andTable3to5.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6173319","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Physical Sciences - Article","associatedPublications":[],"authors":[{"id":466571987,"identity":"b95f83bc-c252-447c-ac5b-d6747fd42079","order_by":0,"name":"Molly Anderson","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0001-7959-0602","institution":"Woods Hole Oceanographic Institution","correspondingAuthor":true,"prefix":"","firstName":"Molly","middleName":"","lastName":"Anderson","suffix":""},{"id":466571988,"identity":"f66f033b-f6ae-4072-ace5-59261c01b476","order_by":1,"name":"Gabriel Johnston","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Gabriel","middleName":"","lastName":"Johnston","suffix":""},{"id":466571989,"identity":"dd50cb63-4fca-4551-a370-9134475b317e","order_by":2,"name":"Michael Perfit","email":"","orcid":"https://orcid.org/0000-0002-1399-8928","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Perfit","suffix":""},{"id":466571990,"identity":"b1bc367e-9781-41d6-9f4c-e43ecce30e6f","order_by":3,"name":"Alessandro Forte","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Alessandro","middleName":"","lastName":"Forte","suffix":""},{"id":466571991,"identity":"cc36cb2e-e4ba-4432-b280-74b0d779b39f","order_by":4,"name":"George Kamenov","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"George","middleName":"","lastName":"Kamenov","suffix":""},{"id":466571992,"identity":"34c89273-5bcb-4a10-aae6-c5d98f4e93da","order_by":5,"name":"David Graham","email":"","orcid":"https://orcid.org/0000-0002-7411-1905","institution":"Oregon State University","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Graham","suffix":""},{"id":466571993,"identity":"25b99bf5-a5a3-41d2-98d6-9aa56802bea9","order_by":6,"name":"Jackie Dixon","email":"","orcid":"","institution":"University of South Florida","correspondingAuthor":false,"prefix":"","firstName":"Jackie","middleName":"","lastName":"Dixon","suffix":""},{"id":466571994,"identity":"63a0cef9-9f4e-4bb8-85c1-7c2c3b8591c2","order_by":7,"name":"V. Dorsey Wanless","email":"","orcid":"","institution":"Woods Hole Oceanographic Institution","correspondingAuthor":false,"prefix":"","firstName":"V.","middleName":"Dorsey","lastName":"Wanless","suffix":""},{"id":466571995,"identity":"2be5d7fb-ae27-41f1-834f-19c3c54ce381","order_by":8,"name":"Petar Glišović","email":"","orcid":"","institution":"Institut de Physique du Globe de Paris","correspondingAuthor":false,"prefix":"","firstName":"Petar","middleName":"","lastName":"Glišović","suffix":""}],"badges":[],"createdAt":"2025-03-06 20:31:04","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6173319/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6173319/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85032207,"identity":"8f7a97c8-c78a-4c7b-978f-a4fae087d0da","added_by":"auto","created_at":"2025-06-20 07:38:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":768916,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eHe, Pb, Sr, and Nd isotopic compositions of the 8°20’ N seamount lavas (triangles) compared with Siqueiros transform lavas (filled blue squares), 9°50’ N on-axis lavas (filled grey circles\u003c/em\u003e\u003csup\u003e\u003cem\u003e38,39\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e), other northern EPR off-axis seamount lavas (filled grey diamonds\u003c/em\u003e\u003csup\u003e\u003cem\u003e16\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e), Galapagos islands (light grey filled squares\u003c/em\u003e\u003csup\u003e\u003cem\u003e40–43\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e), and HIMU-hosted ocean island setting lavas (open symbols\u003c/em\u003e\u003csup\u003e\u003cem\u003e31–37\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e). A) \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb vs \u003c/em\u003e\u003csup\u003e\u003cem\u003e3\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eHe/\u003c/em\u003e\u003csup\u003e\u003cem\u003e4\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eHe (R/R\u003c/em\u003e\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) showing a broad anti-correlation among the northern EPR seamount lavas. Highly differentiated/degassed samples have been filtered out (Figure S2); B) \u003c/em\u003e\u003csup\u003e\u003cem\u003e87\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eSr/\u003c/em\u003e\u003csup\u003e\u003cem\u003e86\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eSr vs \u003c/em\u003e\u003csup\u003e\u003cem\u003e143\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eNd/\u003c/em\u003e\u003csup\u003e\u003cem\u003e144\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eNd; C) \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb vs \u003c/em\u003e\u003csup\u003e\u003cem\u003e87\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eSr/\u003c/em\u003e\u003csup\u003e\u003cem\u003e86\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eSr; D) \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb vs \u003c/em\u003e\u003csup\u003e\u003cem\u003e143\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eNd/\u003c/em\u003e\u003csup\u003e\u003cem\u003e144\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eNd. Uncertainties on Pb, Sr, Nd, and He ratios are smaller than the symbols used. The color scheme for 8°20’ N lavas is as follows: purple symbols are EMORB, light blue are NMORB, and orange are DMORB (categorization based on trace element ratio criteria\u003c/em\u003e\u003csup\u003e\u003cem\u003e13\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e). Two 8°20’ N HIMU (orange triangles) are duplicate analyses on new sample aliquots that were leached in 6 N HCl prior to duplicate Pb, Sr, and Nd isotopic analysis. Estimates of global mantle end-members are labeled for reference\u003c/em\u003e\u003csup\u003e\u003cem\u003e1,3,42,44,45\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/8d6a09e63855a39337d31c7a.png"},{"id":85031440,"identity":"0374e8bc-f4c9-4b07-93bf-c21778d8218b","added_by":"auto","created_at":"2025-06-20 07:30:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":765839,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eHIMU Pb reverse evolutionary models for \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb vs A) \u003c/em\u003e\u003csup\u003e\u003cem\u003e208\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb and B) \u003c/em\u003e\u003csup\u003e\u003cem\u003e207\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb isotopic compositions of the 8°20’ N seamount lavas (filled triangles) and Siqueiros transform (blue filled squares) relative to the NHRL (black solid line). Samples are compared with 9°50’ N on-axis lavas (filled grey circles\u003c/em\u003e\u003csup\u003e\u003cem\u003e38,39\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e), other northern EPR off-axis seamount lavas (filled grey diamonds\u003c/em\u003e\u003csup\u003e\u003cem\u003e16\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e), altered oceanic crust (AOC; open circles) and sediments (circles with a cross) from the Marianas and Izu-Bonin ODP sites\u003c/em\u003e\u003csup\u003e\u003cem\u003e74\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e, as well as an approximate Pb isotope range for Pacific marine sediments\u003c/em\u003e\u003csup\u003e\u003cem\u003e74,75\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. The approximate expected path of evolution for AOC with carbonate veins is indicated with an arrow, which notably deviates from the 8°20’ N seamount HIMU. Two reverse Pb evolution models are calculated starting with a zero-age \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb ratio of 20.2 (current ratio of the 8°20’ N seamount HIMU samples), using the following equations: (1) \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e = \u003c/em\u003e\u003csup\u003e\u003cem\u003e206\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003etoday\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e – µ × (e\u003c/em\u003e\u003csup\u003e\u003cem\u003eλt\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e-1); (2) \u003c/em\u003e\u003csup\u003e\u003cem\u003e208\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e = \u003c/em\u003e\u003csup\u003e\u003cem\u003e208\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003etoday\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e – ω × (e\u003c/em\u003e\u003csup\u003e\u003cem\u003eλt\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e-1; and (3) \u003c/em\u003e\u003csup\u003e\u003cem\u003e207\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e = \u003c/em\u003e\u003csup\u003e\u003cem\u003e207\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb\u003c/em\u003e\u003csub\u003e\u003cem\u003etoday\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e – \u003c/em\u003e\u003csup\u003e\u003cem\u003e235\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eU/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb × (e\u003c/em\u003e\u003csup\u003e\u003cem\u003eλt\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e-1), where t = Time (Ma), λ(\u003c/em\u003e\u003csub\u003e\u003cem\u003e238U\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) = 1.55125x10\u003c/em\u003e\u003csup\u003e\u003cem\u003e-10\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eyr\u003c/em\u003e\u003csup\u003e\u003cem\u003e-1\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e, λ(\u003c/em\u003e\u003csub\u003e\u003cem\u003e232Th\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) = 0.4948x10\u003c/em\u003e\u003csup\u003e\u003cem\u003e-10\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eyr\u003c/em\u003e\u003csup\u003e\u003cem\u003e-1\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e, and λ(\u003c/em\u003e\u003csub\u003e\u003cem\u003e235U\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) = 9.8485x10\u003c/em\u003e\u003csup\u003e\u003cem\u003e-10\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eyr\u003c/em\u003e\u003csup\u003e\u003cem\u003e-1\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. For Model 1 (dotted grey line), µ = 21 based on the current \u003c/em\u003e\u003csup\u003e\u003cem\u003e238\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eU/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb of the 8°20’ N HIMU. For Model 2 (dashed orange line), µ = 60, closer to typical \u003c/em\u003e\u003csup\u003e\u003cem\u003e238\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eU/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb of altered oceanic slabs\u003c/em\u003e\u003csup\u003e\u003cem\u003e27,74\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. For each model, \u003c/em\u003e\u003csup\u003e\u003cem\u003e208\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb is computed using a κ range from 3 to 4. Dashes along the model evolutionary paths denote relevant ages associated with geodynamic particle tracking constraints on recycling age (130 Ma and 240 Ma). The model results indicate the HIMU Pb isotopes must be a mixture of AOC and sediments, and importantly the Pb isotopes must be dominated by sedimentary Pb to produce sufficiently high \u003c/em\u003e\u003csup\u003e\u003cem\u003e207\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb/\u003c/em\u003e\u003csup\u003e\u003cem\u003e204\u003c/em\u003e\u003c/sup\u003e\u003cem\u003ePb in the past 130-240 million years.\u003c/em\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/26d35505b773986dccd87257.png"},{"id":85032209,"identity":"c355b7d6-3ffa-422f-897f-2fc748a3d7a4","added_by":"auto","created_at":"2025-06-20 07:38:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":889554,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e(A) Trace element abundances of 8°20’ N seamount lavas (grey region and black lines) normalized to primitive mantle\u003c/em\u003e\u003csup\u003e\u003cem\u003e4\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e, compared with depleted MORB mantle (DMM), Bulk recycled crust, St Helena HIMU, global subducted sediments (GLOSS), subduction-modified basalt (SMB), a mixture of 1% GLOSS + 99% SMB\u003c/em\u003e\u003csup\u003e\u003cem\u003e27\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e(the mantle component mixture selected here), and a mixture of 2% GLOSS + 98% SMB. (B) Trace element abundances of 8°20’ N HIMU lavas compared with 1% GLOSS + 99% SMB and 2% GLOSS + 98% SMB with error envelopes propagated from GLOSS standard deviation estimates\u003c/em\u003e\u003csup\u003e\u003cem\u003e60\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. Modeled melt and residue compositions using kds from ref\u003c/em\u003e\u003csup\u003e\u003cem\u003e82\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e and assuming modal batch melting represent the two-stage melt model invoked here. Given the geologic context for this seamount chain, we assume two stages of melting, the first normal high-degree melting at the EPR ridge axis, which leaves an incompatible element depleted residue in the mantle which is then rafted off-axis to the approximate location of Coral seamount today. The first stage of melting (20% melt of the mixture of 1% GLOSS + 99% SMB) is extracted to produce a highly depleted residue (grey model, calculated using the modal batch melting equation C_residue = [(D-D*F)/(1-F)]*[Cs/(D+F(1-D)]), and assuming a pyroxenitic lithology containing 30% olivine, 50% orthopyroxene, 20% clinopyroxene, and assuming the same kds from ref\u003c/em\u003e\u003csup\u003e\u003cem\u003e82\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. After the residue has moved away from the ridge, anomalous off-axis melting associated with the seamount chain’s formation generates a second stage of melting of the refractory residue. The second stage melt model therefore simulates a 5 and 10% melt of the residue calculated in the first stage described above. The second stage of melting is calculated using the modal batch melting equation C_melt = Cs/(F+D(1-F)), and assumes deeper melting of a garnet lherzolite (45% ol, 30% opx, 15% cpx, and 10% garnet) over a melt degree range of 5-15% (light and dark blue models which encompass the 8°20’ N HIMU with the exception of Rb).\u003c/em\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/5f96af0a260b6b68e5542547.png"},{"id":85033010,"identity":"ab8b5dc6-5b10-4403-8469-249defc8fbda","added_by":"auto","created_at":"2025-06-20 07:46:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1113496,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eTomography-based reconstructions of time-dependent mantle flowlines and particle track (dark blue line) reversing in time through the mantle over the past 130 million years (panels A, C, D) using V1 viscosity model and methods\u003c/em\u003e\u003csup\u003e\u003cem\u003e63\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. Light blue lines show projection of flow lines onto the surface. Red lines represent current plate boundaries, and black lines outline current continent boundaries. The particle starts from the current position of the 8°20’ N seamounts (D) and reverses directly into the lower mantle (C) to a depth of ~2500 km (~ 387 km above the core-mantle boundary), before returning directly to the upper mantle at ~130 Ma (A) beneath a paleo subduction zone (B) whereby the paleo Mezcalera plate was subducting beneath the eastern edge of the paleo-Farallon plate, forming the Caribbean-Guerrero arc\u003c/em\u003e\u003csup\u003e\u003cem\u003e67\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e. To preserve heterogeneity found across the 8°20’ N seamount lavas, and especially to generate unusually incompatible element depleted HIMU lavas with high U/Pb ratios, multi-stage melting of variable mantle sources may be required (E) where incompatible elements are preferentially extracted in the near-ridge melt region, leaving a refractory yet isotopically heterogeneous mantle off-axis (Stage 1). Subsequently, off-axis melting associated with the anomalous 8°20’ N seamount formation produces incompatible element depleted heterogeneous lavas (Stage 2 melting).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/7f3041e3f2066f3fae221ecd.png"},{"id":85034172,"identity":"fba85a6d-358a-4902-91ab-3ad3a742de2f","added_by":"auto","created_at":"2025-06-20 08:02:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4385775,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/e84c25ce-7514-4cb3-bc63-2a69bed60922.pdf"},{"id":85033807,"identity":"036d2b02-a3d9-4ff6-ab09-f3b76d068c01","added_by":"auto","created_at":"2025-06-20 07:54:19","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":207701,"visible":true,"origin":"","legend":"Supplementary Data Table 1","description":"","filename":"SupplementaryDataTable1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/024bc98f97ef94b14c7e00f7.xlsx"},{"id":85031443,"identity":"e32606c4-2c87-4f2b-9b1b-7dd664379044","added_by":"auto","created_at":"2025-06-20 07:30:19","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":517420,"visible":true,"origin":"","legend":"Supplementary Data Table 2","description":"","filename":"SupplementaryDataTable2Mixingcalculations.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/4dbcb30d44c065b5d10a7f19.xlsx"},{"id":85032211,"identity":"33f466eb-d2b7-411b-ab8f-d209feae5006","added_by":"auto","created_at":"2025-06-20 07:38:19","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":4631922,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialfigure1to8andTable3to5.docx","url":"https://assets-eu.researchsquare.com/files/rs-6173319/v1/f925cdf3108479763a229d07.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Deeply Recycled Origins of Mantle Heterogeneity at the Northern East Pacific Rise","fulltext":[{"header":"Discussion","content":"\u003cp\u003eMid-ocean ridge and ocean island lavas serve as recorders of the time-integrated geochemical evolution of the mantle, tracing the extent and variability of mantle heterogeneities and subduction inputs over time \u003csup\u003e1\u0026ndash;4\u003c/sup\u003e. The full spectrum of mantle source compositions may be obscured, however, in lavas erupted on the spreading ridge-axis, because these melts are aggregated during ascent, and are further homogenized in shallow crustal melt lenses, especially at medium\u0026mdash;fast spreading ridges \u003csup\u003e5\u0026ndash;11\u003c/sup\u003e. Small-scale heterogeneity in the upper mantle is better sampled by examining lavas erupted off-axis \u003csup\u003e1,8,12\u0026ndash;20\u003c/sup\u003e. Thus, off-axis seamounts near the EPR can be used to determine the origin and evolution of the Pacific mantle components over time, and to constrain the geodynamic controls of one of the fastest-spreading ridges on Earth.\u003c/p\u003e \u003cp\u003eAlthough there is no canonical geophysical evidence for mantle plume-derived upwelling directly beneath the northern EPR\u003csup\u003e21\u003c/sup\u003e, some early studies associated widespread geochemical heterogeneity embedded in the sub-EPR mantle, and preferential formation of off-axis seamounts on the Pacific plate, with horizontal flow from plumes distal to the EPR \u003csup\u003e18,22\u003c/sup\u003e. Other studies have invoked long-term injections of recycled oceanic crust, sediments, lithosphere, or some combination of metasomatic recycled components into the mantle \u003csup\u003e1,3,15,16,18,20,23\u0026ndash;28\u003c/sup\u003e. Regardless of the nature of the subducted component, geochemical models alone provide limited constraints on spatiotemporal origins of mantle heterogeneities. Recent time-dependent mantle flow predictions from seismic tomography reveal a broad, deep-seated, mantle-wide upwelling beneath the EPR\u003csup\u003e29,30\u003c/sup\u003e, which could provide a solution to the enigmatic origin of EPR mantle heterogeneity.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe 8\u0026deg;20\u0026rsquo; N seamount lavas exhibit extreme heterogeneity among major and trace element concentrations and radiogenic isotopes over short (sub-km) length scales (e.g., Figure S1), providing new constraints on northern EPR mantle heterogeneity\u003csup\u003e13\u003c/sup\u003e. The major and trace element concentrations, Pb, Sr, Nd and He isotope ratios are reported in Table S1. The compositional range of the full suite of seamount lavas spans the entire range \u0026ndash; and in some cases extends beyond \u0026ndash; previously defined values of the northern EPR MORB and its adjacent seamounts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e). 8\u0026deg;20\u0026rsquo; N seamount [La/Sm]\u003csub\u003eN\u003c/sub\u003e = 0.30 to 2.85 (northern EPR and adjacent seamounts\u0026thinsp;=\u0026thinsp;0.22 to 3.4). In addition, \u003csup\u003e3\u003c/sup\u003eHe/\u003csup\u003e4\u003c/sup\u003eHe ranges from 6.4\u0026ndash;9.2 R\u003csub\u003eA\u003c/sub\u003e (where R\u003csub\u003eA\u003c/sub\u003e is the atmospheric ratio), and is broadly anti-correlated with \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb, which ranges from 17.5\u0026ndash;20.2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e). By contrast, the northern EPR and other adjacent seamounts have \u003csup\u003e3\u003c/sup\u003eHe/\u003csup\u003e4\u003c/sup\u003eHe up to 8.7 R\u003csub\u003eA\u003c/sub\u003e and \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb = 18.2\u0026ndash;19.8\u003csup\u003e16\u003c/sup\u003e (Figure S2). Importantly, several samples from one seamount (Coral) define a new end-member component with HIMU characteristics, possessing high \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb ~\u0026thinsp;20.2 and \u003csup\u003e208\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb ~\u0026thinsp;40.2. This HIMU component is similar to that found in the French Polynesian islands\u003csup\u003e31\u0026ndash;36\u003c/sup\u003e and at St. Helena\u003csup\u003e37\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e), but has not previously been found at the EPR.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Nature of the Multi-Component Mantle Feeding the Northern EPR","content":"\u003cp\u003eCombined radiogenic He, Pb, Sr, and Nd isotopes form an important framework for discriminating between oceanic mantle sources. Because radiogenic isotopes do not fractionate during shallow-level or recent magma evolutionary processes, their ratios record mantle source trace element ratios (e.g., U/Pb, U\u0026thinsp;+\u0026thinsp;Th/He, Rb/Sr, and Sm/Nd) integrated over geologic time. High-density sampling of oceanic crust especially at off-axis seamounts reveals the EPR mantle is isotopically highly heterogeneous \u003csup\u003e13,16,18,19,47\u0026ndash;50\u003c/sup\u003e. Some isotopically enriched but incompatible element depleted lavas erupt along the EPR crest \u003csup\u003e10,51,52\u003c/sup\u003e, although these compositions are mostly found off-axis. Some studies have invoked two mantle components in the northern sub-EPR mantle: an incompatible element depleted MORB mantle (DMM) and an incompatible element enriched source (EMM) \u003csup\u003e13,38,52\u0026ndash;54\u003c/sup\u003e. Other studies have argued for at least three mantle components in the near-EPR mantle \u003csup\u003e15,23\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur results show that the sub-EPR mantle in a region far from a mantle plume contains at least four end-member mantle components: 1) an incompatible element (Nb/La\u0026thinsp;\u0026lt;\u0026thinsp;1) and isotopically depleted source (DMM; \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb \u0026lt; 18.5); 2) an incompatible element depleted (Nb/La\u0026thinsp;\u0026lt;\u0026thinsp;0.8) HIMU source (\u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb \u0026gt; 20.1); 3) an incompatible element enriched (Nb/La\u0026thinsp;\u0026gt;\u0026thinsp;1.5), isotopically enriched source (possibly EMII); and 4) an incompatible element enriched source with isotopic compositions similar to EMI (radiogenic Sr and Nd, and unradiogenic Pb). The discovery of HIMU lavas at the northern EPR is consistent with the composition of several off-axis seamounts near the EPR approaching HIMU characteristics \u003csup\u003e8,16\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Traditionally mid-ocean ridge heterogeneities (all but DMM) are associated with enrichment of incompatible elements, but in this study depleted MORB (DMORB) alone span the entire Pb, Nd, and He isotopic range, indicating the depleted mantle is heterogeneous\u003csup\u003e17\u003c/sup\u003e. Commonly invoked models accounting for origins of mantle heterogeneity include injections of recycled oceanic crust and slab derived fluids/melts \u003csup\u003e15,20,35,36,55\u0026ndash;58\u003c/sup\u003e, recycled sediments \u003csup\u003e15,35\u003c/sup\u003e, recycled metasomatized lithosphere \u003csup\u003e4,28,37\u003c/sup\u003e and recycled carbonates \u003csup\u003e24,25,59\u003c/sup\u003e. Further, the timing and location of subduction, and the origin and nature of recycled components in the EPR mantle have remained unclear due to the inherently heterogeneous nature of globally subducted lithosphere and sediments\u003csup\u003e60\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eActive Upwelling Mantle Source for Heterogeneity Beneath the EPR\u003c/h2\u003e \u003cp\u003eFor the first time, we investigate the origins of EPR mantle heterogeneities identified among the 8\u0026deg;20\u0026rsquo; N seamount lavas (the HIMU end-member specifically) by using tomography-based time-reversed mantle convection models\u003csup\u003e61,62\u003c/sup\u003e. To track the time-reversed convective evolution of parcels of the mantle, seismic tomography images of present-day mantle structures are used as starting conditions\u003csup\u003e63,64\u003c/sup\u003e. For tomographically well-characterized regions like the Pacific, present-day tomographic models can be used to infer past mantle structures and mantle flow, through time-reversed reconstructions of the evolution of mantle convection using the Back-and-Forth Nudging (BFN) numerical method employed here\u003csup\u003e61,63\u003c/sup\u003e. Surface boundary conditions are provided by coupling the motions of rigid plates (using constraints from plate reconstructions) with the underlying mantle flow, and two depth-dependent viscosity profiles (V1 and V2) are considered \u003csup\u003e65,66\u003c/sup\u003e. The resulting reconstructed mantle flowlines (particle tracks) are shown relative to current plate orientations at the surface (Figure S3; Figure S4).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe V1 viscosity mantle flow model\u003csup\u003e65\u003c/sup\u003e is favored here due to the thinner, lower-viscosity lithosphere and shallow asthenospheric layer, which make it ideal for environments like mid-ocean ridges characterized by thinner lithosphere. Our particle track simulation, based on the V1 mantle viscosity, is initialized at 8.409454 N and \u0026minus;\u0026thinsp;104.412 E, and starts 50 km below the present-day location of the 8\u0026deg;20\u0026rsquo; N seamounts. The time-reversed particle trajectory reaches its maximum depth of ~\u0026thinsp;2500 km (or ~\u0026thinsp;387 km above the core-mantle boundary) within ~\u0026thinsp;76 Ma (Figure S3). The particle then returns to the upper mantle around \u0026minus;\u0026thinsp;85\u0026deg; E, 27\u0026deg; N, placing a time constraint on subduction-driven entry into the convecting lower mantle at around 130 Ma. Taking into account slab descent times (between ~\u0026thinsp;10 and ~\u0026thinsp;20 Ma) across the upper mantle\u003csup\u003e68\u003c/sup\u003e, these results are consistent with independently constrained plate-history reconstructions that suggest the paleo-Mezcalera plate was subducting beneath the eastern edge of the paleo-Farallon plate forming the Caribbean-Guerrero arc\u003csup\u003e67\u003c/sup\u003e. Prior time-dependent tomographic mantle flow models have predicted deep-sourced active upwelling beneath the EPR\u003csup\u003e62\u003c/sup\u003e, the results from which are consistent with the relative stability/immobility of the EPR over the past 90\u0026nbsp;million years despite asymmetric far-field slab pull\u003csup\u003e29,30\u003c/sup\u003e, as is indicated by the new particle tracking models.\u003c/p\u003e \u003cp\u003eBased on particle tracking simulations, we consider two possibilities for the origin of EPR mantle heterogeneity: (1) rapid, full-mantle convection has recently dredged up recycled material that has potentially been residing in the lower mantle as deep as 400 km above the core-mantle boundary, in which case these recycling models constrain the timescales for elemental transport from source to sink across the length of the mantle; or, (2) Early Cretaceous material from a paleo subduction zone was rapidly convected through the mantle to ~\u0026thinsp;400 km above the core-mantle boundary, and returned to the upper mantle beneath the EPR over the past ~\u0026thinsp;130 to ~\u0026thinsp;150\u0026nbsp;million years, taking into account possible subduction transit times across the upper mantle\u003csup\u003e68\u003c/sup\u003e. At this time, we cannot rule out the possibility of recently-dredged, ancient, deep-mantle material feeding the EPR. However, the extent of mixing along the particle track inferred appears to be limited (supplementary discussion C). If the latter case is true, we can provide new lower-limit constraints on elemental cycling through the mantle. We combine radiogenic isotope evolution models with trace element mixing-melting models to demonstrate how recently subducted material could feasibly produce heterogeneity present in the mantle beneath the EPR today.\u003c/p\u003e \u003c/div\u003e"},{"header":"Recent Recycling Origins for EPR Mantle Heterogeneity","content":"\u003cp\u003eAssuming the EPR heterogeneity currently beneath the ridge has contributions from Mezcalera-Farallon subduction-related processes, the composition of the material subducting beneath the paleo-Farallon plate between ~\u0026thinsp;150 and ~\u0026thinsp;130\u0026nbsp;million years ago can be assessed by comparing radiogenic isotope ratios of the 8\u0026deg;20\u0026rsquo; N seamount lavas with their trace element compositions. For the isotopic evolutionary model, we focus on the HIMU component because the 8\u0026deg;20\u0026rsquo; N seamount HIMU closely resembles the end-member isotope ratios of St Helena and French Polynesia, whereas the alternative recycling-related components (EM) appear to be mixtures with DMM and therefore make it more difficult to determine a time frame of mixing that distinguishes a mantle end-member (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHIMU lavas are characterized by highly radiogenic \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb (paired with unradiogenic \u003csup\u003e87\u003c/sup\u003eSr/\u003csup\u003e86\u003c/sup\u003eSr), requiring high time-integrated U/Pb and Th/Pb but low Rb/Sr to generate extreme enrichment in \u003csup\u003e206\u003c/sup\u003ePb and \u003csup\u003e208\u003c/sup\u003ePb isotopes over time paired with only moderate enrichment in \u003csup\u003e87\u003c/sup\u003eSr\u003csup\u003e1\u003c/sup\u003e. A widely favored model involves HIMU originating from ancient recycled oceanic crust which was hydrothermally altered before subduction and reintroduced to the mantle between 2.0\u0026ndash;3.0 Ga\u003csup\u003e2,33,36\u003c/sup\u003e. This model is largely reliant on Pb extraction during subduction which would result in high U/Pb ratios in the altered crustal precursors integrated over billions of years during recycling. Sulfur isotopes further constrain recycling ages for some ocean island HIMU sources to the Archean\u003csup\u003e69\u0026ndash;71\u003c/sup\u003e. Given the new mantle convection constraints presented here, this type of origin for the northern EPR HIMU would be possible if ancient reservoirs previously trapped in the deep mantle are being sampled recently by whole-mantle convection beneath the EPR.\u003c/p\u003e \u003cp\u003eHIMU lavas also typically possess \u003csup\u003e3\u003c/sup\u003eHe/\u003csup\u003e4\u003c/sup\u003eHe values lower than typical MORB (7\u0026ndash;10 R\u003csub\u003eA\u003c/sub\u003e), suggesting materials forming HIMU have high time-integrated U\u0026thinsp;+\u0026thinsp;Th/\u003csup\u003e3\u003c/sup\u003eHe relative to MORB\u003csup\u003e16,44,45\u003c/sup\u003e. However, on billion-year timescales, substantially more \u003csup\u003e4\u003c/sup\u003eHe should have accumulated\u003csup\u003e35\u003c/sup\u003e. The decoupling of U/Pb from (U\u0026thinsp;+\u0026thinsp;Th)/He poses a problem for ancient altered subducted oceanic crustal origins of HIMU\u003csup\u003e36,72,73\u003c/sup\u003e. Several models potentially explain the relative uniformity of \u003csup\u003e3\u003c/sup\u003eHe/\u003csup\u003e4\u003c/sup\u003eHe values among ocean island HIMU: (1) open-system behavior of helium due to its high diffusivity and equilibration with surrounding mantle\u003csup\u003e35\u003c/sup\u003e, (2) incorporation of relatively less-degassed mantle into OIB HIMU sources\u003csup\u003e32\u003c/sup\u003e, (3) significantly lower (U\u0026thinsp;+\u0026thinsp;Th)/He than anticipated from old subducted oceanic crust (i.e., the addition of carbonates to subducting crust), or (4) timescales of recycling to form some HIMU sources are significantly shorter than traditionally thought.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe predicted mantle-flow velocity field (\u0026ldquo;mantle wind\u0026rdquo;) that is reflected in the particle trajectories requires the consideration of a relatively recent origin for EPR HIMU. If \u0026micro; is sufficiently high upon subduction, HIMU \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb ratios can be produced in \u0026lt;\u0026thinsp;300\u0026nbsp;million years\u003csup\u003e76,77\u003c/sup\u003e. For instance, the presence of carbonate veins in altered oceanic crust can significantly increase \u0026micro;, or hydrothermal alteration on the seafloor increases the U/Pb of oceanic crust by preferentially leaching Pb\u003csup\u003e74,76\u003c/sup\u003e. To assess the possibility of a young origin of the 8\u0026deg;20\u0026rsquo; N seamount HIMU, reverse Pb evolution models are presented (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Figure S5), starting with the 8\u0026deg;20\u0026rsquo; N seamount HIMU present-day Pb isotope ratios, \u0026micro; (\u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e204\u003c/sup\u003ePb) of 21\u0026ndash;60, and κ of 3\u0026ndash;4. Although the EPR HIMU could be produced in ~\u0026thinsp;240 Ma by extremely radiogenic Pb isotope ratios consistent with Pacific sediments, we favor a model where \u0026micro; is closer to 60, consistent with old altered oceanic slabs containing carbonate veins \u003csup\u003e27,74\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In this case, the northern EPR HIMU could have formed in the past 130 Ma, consistent with the V1 particle tracking model if the Pb isotopes are initially dominated by the presence of sediments (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Even if the mixture involved only small amounts of sediment (for example, 1\u0026ndash;2 %) and 98\u0026ndash;99% ceanic crust (Table S2), Pb isotopes would be dominated by sediments (e.g., GLOSS Pb\u0026thinsp;~\u0026thinsp;20 ppm whereas MORB Pb\u0026thinsp;~\u0026thinsp;0.48 ppm). By contrast, Sr and Nd would be impacted to a lesser extent (sediment Sr\u0026thinsp;~\u0026thinsp;327 ppm whereas MORB Sr\u0026thinsp;~\u0026thinsp;136 ppm; sediment Nd\u0026thinsp;~\u0026thinsp;27 ppm whereas MORB Nd\u0026thinsp;~\u0026thinsp;10 ppm)\u003csup\u003e60,78\u003c/sup\u003e, which would explain why the Sr and Nd isotope ratios of the HIMU lavas do not imply a sedimentary origin.\u003c/p\u003e \u003cp\u003eThe Pb, Sr, Nd, and He isotopes and trace element ratios of the seamount HIMU basalts are consistent with recycled oceanic slab origins, but the trace element concentrations are different from typical incompatible element enriched HIMU basalts such as those from St Helena (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003e). For instance, although the measured U/Pb for the seamount HIMU lavas (\u0026micro;\u0026thinsp;~\u0026thinsp;21) is high compared to normal depleted mantle (\u0026micro;\u0026thinsp;~\u0026thinsp;5), the most incompatible trace elements are extremely depleted (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003e). High Rb/Sr and U/Pb are expected to result from hydrothermal alteration of young crust near the ridge, although this alteration process alone is not sufficient to create HIMU without additional alteration upon subsequent subduction \u003csup\u003e20,79\u003c/sup\u003e. Subduction related dehydration processes typically are expected to remove Pb, Sr, Rb (Rb more than Sr), Ca, Ba, K, and light rare earth elements (e.g., Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003eA), also resulting in high U/Pb and intermediate Rb/Sr\u003csup\u003e15\u003c/sup\u003e. Despite having trace element ratio characteristics of subducted oceanic crust, the seamount HIMU primitive mantle-normalized trace element compositions are more depleted than bulk recycled crust, unlike typical ocean island HIMU lavas like those found at St. Helena (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Variable degrees of melting alone cannot explain differences in trace element abundances between EPR seamount HIMU and ocean island HIMU (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Either the nature of the EPR HIMU source is different from (and more depleted than) plume-sourced HIMU, or melting, mixing, and metasomatic processes drastically differ between these settings. For instance, the difference may be attributed to the degree to which the peridotite dominated mantle is contaminated and metasomatized; the mid-ocean ridge mantle being peridotite-dominated, whereas ocean island plume mantle melts can be derived from blended peridotite-pyroxenite lithologies in which the enriched lithologies metasomatically induce trace element enrichment of surrounding peridotite. This ultimately masks the truly incompatible element depleted nature of the HIMU mantle source. If HIMU sources at ocean islands are typically dominated by incompatible element depleted oceanic crust, then we would still expect to see depletion in incompatible element concentrations among HIMU basalts worldwide\u003csup\u003e1\u003c/sup\u003e. Thus, we argue that the 8\u0026deg;20\u0026rsquo; N seamount HIMU may be representative of primary, incompatible element depleted HIMU mantle sources whereas ocean island HIMU have been overprinted by incipient melts of enriched lithologies embedded in mantle peridotite.\u003c/p\u003e \u003cp\u003eTo produce the trace element patterns of the northern EPR HIMU consistent with a mixture of subduction-modified basalt and sediments (as inferred from the Pb evolution models), multi-stage melting may be required (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, \u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e4\u003c/span\u003e) \u003csup\u003e80,81\u003c/sup\u003e. In this scenario, a mixture of 1% GLOSS and 99% subduction-modified basalt is first melted beneath the EPR in the early stages of the seamount chain\u0026rsquo;s development (Stage 1). After the refractory residue from ridge melting has rafted off-axis, a second stage of melting off-axis melting responsible for forming the seamounts produces a 5\u0026ndash;15% partial melt (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e4\u003c/span\u003e) consistent with the HIMU trace element patterns observed at Coral seamount, such as extreme Pb, Ba, and Th depletion (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Alternatively, the highly depleted EPR HIMU trace element pattern could be formed by melting a mixture of 95% DMM with 5% recycled material (ie, the 1% GLOSS\u0026thinsp;+\u0026thinsp;99% sediment mixture) (Figure S6). However, in this simple (batch melting) scenario, the off-axis HIMU melts are derived by ~\u0026thinsp;20% partial melting, which may be unrealistic off-axis. Regardless of the detailed melting-mixture scenario, there are multiple ways to create incompatible element depleted HIMU sources given a starting mantle source that includes 1% GLOSS and 99% subduction-modified basalt inferred from Pb isotope models.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOur study provides a new multidisciplinary solution to the long-standing problem concerning the origin of EPR mantle heterogeneity. Combined geochemical and geodynamic models imply relatively short timescales for recycling deep into the mantle followed by rapid upwelling today beneath the EPR. Lavas from the 8\u0026deg;20\u0026rsquo; N seamount chain record greater geochemical heterogeneity than has been observed along axis, including evidence for a 4-component mantle, and the first discovery of a high-\u0026micro; mantle source at the northern EPR. This HIMU component is uniquely incompatible element depleted compared with global ocean island HIMU, potentially representing the primary depleted nature of HIMU sources worldwide if they derive from recycled lithosphere. The combined models tie the origin of current sub-EPR mantle heterogeneity to a paleo subduction zone at ~\u0026thinsp;130 to ~\u0026thinsp;150 Ma. Geochemical, geodynamic, and plate reconstruction models are independently consistent with recently subducted, deeply recycled mantle heterogeneities feeding the northern EPR, which may mean that deep-seated mantle buoyancy drives the fastest spreading ridge on Earth. A full consideration of rapid convection timescales is needed in mass balance calculations for the evolution of the Pacific mantle and its role in global elemental cycling. Other localities along the EPR, Pacific-Antarctic Ridge, and Southwest Indian Ridge consist of highly heterogeneous lavas in locations distal to mantle plumes\u003csup\u003e54,83,84\u003c/sup\u003e. In such cases, heterogeneity has been attributed to melting of a marble-cake upper mantle style mixture, but new methods like those in this study have the potential to trace the surface origin of such mantle heterogeneities on a planetary scale.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the captains and crews of the R/V \u003cem\u003eAtlantis\u003c/em\u003e (AT37-05 and AT42-06), the HOV \u003cem\u003eAlvin\u003c/em\u003e and AUV \u003cem\u003eSentry\u003c/em\u003e teams for ensuring successful field operations. We are grateful to the other science team members of the OASIS I and II expeditions (Off-Axis Seamount Investigation at Siqueiros) including P. Gregg, D. Fornari, M. Smith, D. Geist, S. Shirey, C. Lundstrom, B. John, R. Parnell-Turner, H. Cabaniss, B. Boulahanis, E. McCully, C. Trim, V. Romano, Y.J. Tan, J. Albright, and Y. Zhan. We thank M. Lytle and K. Bermudez for invaluable analytical assistance, and greatly appreciate discussions with Veronique Le Roux, Matthew Leybourne, Stephen Elardo, Peter Barry, and Forrest Horton. This work was supported by the National Science Foundation (NSF OCE-MGG 1356610, NSF OCE-MGG 1356822, NSF OCE-MGG 1357150, NSF OCE-MGG 2001314), the Burnham Research Grant at Boise State University, Geological Society of America Lipman Award, and the University of Florida Graduate School Funding Fellowship. Helium isotope analyses at OSU were supported by NSF OCE-MGG1763255.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eData Availability Statement:\u0026nbsp;\u003c/strong\u003eData in support of this manuscript are available in supplemental data and the original data from the publication of 2016 OASIS I cruise samples are online at EarthChem https://doi.org/10.26022/IEDA/111616\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAuthor Contributions:\u0026nbsp;\u003c/strong\u003eAll coauthors assisted with manuscript revision and interpretation of results. MKA collected the samples, performed the laboratory experiments, created the mixing, isotope evolution, and melting models, conceptualized the final model, and wrote the initial manuscript draft with input from coauthors; GJ created and interpreted the geodynamic simulations, and conceptualized the final model; MRP conceived and supervised the study and collected the samples; AF created and interpreted the geodynamic simulations and supervised the study; GK performed the laboratory experiments and created mixing models; DWG performed the laboratory experiments and assisted with isotope evolution and melting models, JD assisted with model calculations, and provided key feedback for coordinating geochemical and geodynamic model results, VDW designed and supervised the study and collected the samples; PG designed the geodynamic models.\u0026nbsp;\u003c/p\u003e"},{"header":"References ","content":"\u003col\u003e\n\u003cli\u003eZindler, A. \u0026amp; Hart, S. Chemical Geodynamics. \u003cem\u003eAnnual Reviews Earth Planetary Science\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 493\u0026ndash;571 (1986).\u003c/li\u003e\n\u003cli\u003eHofmann, A. W. Mantle geochemistry: the message from oceanic volcanism. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e385\u003c/strong\u003e, 219\u0026ndash;229 (1997).\u003c/li\u003e\n\u003cli\u003eStracke, A., Hofmann, A. W. \u0026amp; Hart, S. R. FOZO, HIMU, and the rest of the mantle zoo. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1\u0026ndash;20 (2005).\u003c/li\u003e\n\u003cli\u003eSun, S. S. \u0026amp; McDonough, W. F. Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. \u003cem\u003eGeol Soc Spec Publ\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 313\u0026ndash;345 (1989).\u003c/li\u003e\n\u003cli\u003eHebert, L. B. \u0026amp; Mont\u0026eacute;si, L. G. J. Generation of permeability barriers during melt extraction at mid-ocean ridges. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;17 (2010).\u003c/li\u003e\n\u003cli\u003eKatz, R. F. \u0026amp; Weatherley, S. M. Consequences of mantle heterogeneity for melt extraction at mid-ocean ridges. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e335\u0026ndash;336\u003c/strong\u003e, 226\u0026ndash;237 (2012).\u003c/li\u003e\n\u003cli\u003eKeller, T., Katz, R. F. \u0026amp; Hirschmann, M. M. Volatiles beneath mid-ocean ridges: Deep melting, channelised transport, focusing, and metasomatism. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e464\u003c/strong\u003e, 55\u0026ndash;68 (2017).\u003c/li\u003e\n\u003cli\u003eNiu, Y., Regelous, M., Wendt, I. J., Batiza, R. \u0026amp; O\u0026rsquo;hara, M. J. Geochemistry of near-EPR seamounts: importance of source vs. process and the origin of enriched mantle component. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e199\u003c/strong\u003e, 327\u0026ndash;345 (2002).\u003c/li\u003e\n\u003cli\u003ePhipps Morgan, J. Melt Migration Beneath Mid-Ocean Spreading Centers. \u003cem\u003eGeophys Res Lett\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1238\u0026ndash;1241 (1987).\u003c/li\u003e\n\u003cli\u003ePerfit, M. R. \u0026amp; Chadwick, W. W. Magmatism at mid-ocean ridges: Constraints from volcanological and geochemical investigations. in \u003cem\u003eGeophysical Monograph Series\u003c/em\u003e vol. 106 59\u0026ndash;115 (Blackwell Publishing Ltd, 1998).\u003c/li\u003e\n\u003cli\u003eSinton, J. M. \u0026amp; Detrick, R. S. Mid-ocean ridge magma chambers. \u003cem\u003eJ Geophys Res\u003c/em\u003e \u003cstrong\u003e97\u003c/strong\u003e, 197\u0026ndash;216 (1992).\u003c/li\u003e\n\u003cli\u003eAllan, J. F., Batiza, R. \u0026amp; Lonsdale, P. Petrology and Chemistry of Lavas from Seamounts Flanking the East Pacific Rise Axis, 21 deg N: Implications Concerning the Mantle Source Composition for Both Seamount and Adjacent EPR Lavas. \u003cem\u003eAmerican Geophysical Union Geophysical Monogram Series\u003c/em\u003e 255\u0026ndash;282 (1987).\u003c/li\u003e\n\u003cli\u003eAnderson, M. \u003cem\u003eet al.\u003c/em\u003e Extreme Heterogeneity in Mid-Ocean Ridge Mantle Revealed in Lavas From the 8\u0026deg;20\u0026prime;N Near-Axis Seamount Chain. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eBatiza, R. \u0026amp; Vanko, D. Petrology of young Pacific seamounts. \u003cem\u003eJ Geophys Res\u003c/em\u003e \u003cstrong\u003e89\u003c/strong\u003e, 11235\u0026ndash;11260 (1984).\u003c/li\u003e\n\u003cli\u003eDixon, J. E. \u003cem\u003eet al.\u003c/em\u003e Light Stable Isotopic Compositions of Enriched Mantle Sources: Resolving the Dehydration Paradox. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 3801\u0026ndash;3839 (2017).\u003c/li\u003e\n\u003cli\u003eGraham, D. W. \u003cem\u003eet al.\u003c/em\u003e He, Pb, Sr and Nd isotope constraints on magma genesis and mantle heterogeneity beneath young Pacific seamounts. \u003cem\u003eContrib Mineral Petrol\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 446\u0026ndash;463 (1988).\u003c/li\u003e\n\u003cli\u003eLambart, S. \u003cem\u003eet al.\u003c/em\u003e Highly heterogeneous depleted mantle recorded in the lower oceanic crust. \u003cem\u003eNat Geosci\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 482\u0026ndash;486 (2019).\u003c/li\u003e\n\u003cli\u003eNiu, Y. \u0026amp; Batiza, R. Trace element evidence from seamounts for recycled oceanic crust in the Eastern Pacific mantle. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e148\u003c/strong\u003e, 471\u0026ndash;483 (1997).\u003c/li\u003e\n\u003cli\u003eZindler, A., Staudigel, H. \u0026amp; Batiza, R. Isotope and trace element geochemistry of young Pacific seamounts: implications for the scale of upper mantle heterogeneity. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 175\u0026ndash;195 (1984).\u003c/li\u003e\n\u003cli\u003eDixon, J. E., Leist, L., Langmuir, C. \u0026amp; Schilling, J.-G. Recycled dehydrated lithosphere observed in plume-influenced mid-ocean-ridge basalt. \u003cstrong\u003e420\u003c/strong\u003e, 385\u0026ndash;389 (2002).\u003c/li\u003e\n\u003cli\u003eMeyzen, C. M. \u003cem\u003eet al.\u003c/em\u003e Isotopic portrayal of the Earth\u0026rsquo;s upper mantle flow field. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e447\u003c/strong\u003e, 1069\u0026ndash;1074 (2007).\u003c/li\u003e\n\u003cli\u003ePhipps Morgan, J., Morgan, W. J., Yu-Shen Zhang \u0026amp; Smith, W. H. F. Observational hints for a plume-fed, suboceanic asthenosphere and its role in mantle convection. \u003cem\u003eJ Geophys Res\u003c/em\u003e \u003cstrong\u003e100\u003c/strong\u003e, 12753\u0026ndash;12767 (1995).\u003c/li\u003e\n\u003cli\u003eCastillo, P. R. \u003cem\u003eet al.\u003c/em\u003e Petrology and Sr, Nd, and Pb isotope geochemistry of mid-ocean ridge basalt glasses from the 11\u0026deg;45\u0026prime;N to 15\u0026deg;00\u0026prime;N segment of the East Pacific Rise. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 1\u0026ndash;40 (2000).\u003c/li\u003e\n\u003cli\u003eCastillo, P. R., MacIsaac, C., Perry, S. \u0026amp; Veizer, J. Marine Carbonates in the Mantle Source of Oceanic Basalts: Pb Isotopic Constraints. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1\u0026ndash;7 (2018).\u003c/li\u003e\n\u003cli\u003eCastillo, P. R. The recycling of marine carbonates and sources of HIMU and FOZO ocean island basalts. \u003cem\u003eLithos\u003c/em\u003e \u003cstrong\u003e216\u0026ndash;217\u003c/strong\u003e, 254\u0026ndash;263 (2015).\u003c/li\u003e\n\u003cli\u003eDonnelly, K. E., Goldstein, S. L., Langmuir, C. H. \u0026amp; Spiegelman, M. Origin of enriched ocean ridge basalts and implications for mantle dynamics. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e226\u003c/strong\u003e, 347\u0026ndash;366 (2004).\u003c/li\u003e\n\u003cli\u003eStracke, A., Bizimis, M. \u0026amp; Salters, V. J. M. Recycling oceanic crust: Quantitative constraints. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1\u0026ndash;33 (2003).\u003c/li\u003e\n\u003cli\u003eWeiss, Y., Class, C., Goldstein, S. L. \u0026amp; Hanyu, T. Key new pieces of the HIMU puzzle from olivines and diamond inclusions. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e537\u003c/strong\u003e, 666\u0026ndash;670 (2016).\u003c/li\u003e\n\u003cli\u003eRowley, D. B. \u003cem\u003eet al.\u003c/em\u003e Kinematics and dynamics of the East Pacific Rise linked to a stable, deep-mantle upwelling. \u003cem\u003eSci Adv\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1\u0026ndash;18 (2016).\u003c/li\u003e\n\u003cli\u003eRowley, D. B. \u0026amp; Forte, A. M. Kinematics of the East Pacific Rise Retrodicted From Pacific and Farallon/Nazca Subduction-Related Torques: Support for Significant Deep Mantle Buoyancy Controlling EPR Spreading. \u003cem\u003eJ Geophys Res Solid Earth\u003c/em\u003e \u003cstrong\u003e127\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eLassiter, J. C., Blichert-Toft, J., Hauri, E. H. \u0026amp; Barsczus, H. G. Isotope and trace element variations in lavas from Raivavae and Rapa, Cook-Austral islands: Constraints on the nature of HIMU- and EM-mantle and the origin of mid-plate volcanism in French Polynesia. \u003cem\u003eChem Geol\u003c/em\u003e \u003cstrong\u003e202\u003c/strong\u003e, 115\u0026ndash;138 (2003).\u003c/li\u003e\n\u003cli\u003eParai, R., Mukhopadhyay, S. \u0026amp; Lassiter, J. C. New constraints on the HIMU mantle from neon and helium isotopic compositions of basalts from the Cook-Austral Islands. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e277\u003c/strong\u003e, 253\u0026ndash;261 (2009).\u003c/li\u003e\n\u003cli\u003eHauri, E. H. \u0026amp; Hart, S. R. Re-Os isotope systematics of HIMU and EMII oceanic island basalts from the south Pacific Ocean. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 353\u0026ndash;371 (1993).\u003c/li\u003e\n\u003cli\u003eHanyu, T. \u0026amp; Nakamura, E. Constraints on HIMU and EM by Sr and Nd isotopes re-examined. \u003cem\u003eEarth Planets Space\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 61\u0026ndash;70 (2000).\u003c/li\u003e\n\u003cli\u003eHanyu, T. \u0026amp; Kaneoka, I. The uniform and low 3He/4He ratios of HIMU basalts as evidence for their origin as recycled materials. \u003cem\u003eLetters to Nature\u003c/em\u003e \u003cstrong\u003e390\u003c/strong\u003e, 273\u0026ndash;276 (1997).\u003c/li\u003e\n\u003cli\u003eChauvel, C., Hofmann, A. W. \u0026amp; Vidal, P. HIMU-EM: The French Polynesian Connection. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e110\u003c/strong\u003e, 99\u0026ndash;119 (1992).\u003c/li\u003e\n\u003cli\u003eGraham, D. W., Humphris, S. E., Jenkins, W. J. \u0026amp; Kurz, M. D. Helium isotope geochemistry of some volcanic rocks from Saint Helena. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e110\u003c/strong\u003e, 121\u0026ndash;131 (1992).\u003c/li\u003e\n\u003cli\u003eGoss, A. R. \u003cem\u003eet al.\u003c/em\u003e Geochemistry of lavas from the 2005-2006 eruption at the East Pacific Rise, 9\u0026deg;46\u0026rsquo;N-9\u0026deg;56N: Implications for ridge crest plumbing and decadal changes in magma chamber compositions. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;35 (2010).\u003c/li\u003e\n\u003cli\u003eWaters, C. L. \u003cem\u003eet al.\u003c/em\u003e Sill to surface: Linking young off-axis volcanism with subsurface melt at the overlapping spreading center at 9\u0026deg;03\u0026rsquo;N East Pacific Rise. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e369\u0026ndash;370\u003c/strong\u003e, 59\u0026ndash;70 (2013).\u003c/li\u003e\n\u003cli\u003eGraham, D. W., Christie, D. M., Harpp, K. S. \u0026amp; Lupton, J. E. Mantle Plume Helium in Submarine Basalts from the Galapagos Platform. \u003cem\u003eScience Reports\u003c/em\u003e \u003cstrong\u003e262\u003c/strong\u003e, 2023\u0026ndash;2026 (1993).\u003c/li\u003e\n\u003cli\u003eHarpp, K. S. \u0026amp; Weis, D. Insights Into the Origins and Compositions of Mantle Plumes: A Comparison of Gal\u0026aacute;pagos and Hawai\u0026rsquo;i. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eHarpp, K. S. \u0026amp; White, W. M. Tracing a mantle plume: Isotopic and trace element variations of Gal\u0026aacute;pagos seamounts. \u003cem\u003eGeochem. Geophys. Geosyst\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, (2001).\u003c/li\u003e\n\u003cli\u003eKurz, M. D. \u0026amp; Geist, D. Dynamics of the Galapagos hotspot from helium isotope geochemistry. \u003cem\u003eGeochim Cosmochim Acta\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, (1999).\u003c/li\u003e\n\u003cli\u003eKurz, M. D., Jenkins, W. J. \u0026amp; Hart, S. R. Helium isotopic systematics of oceanic islands and mantle heterogeneity. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e297\u003c/strong\u003e, 97\u0026ndash;99 (1982).\u003c/li\u003e\n\u003cli\u003eGraham, D. W. Noble Gas Isotope Geochemistry of Mid-Ocean Ridge and Ocean Island Basalts: Characterization of Mantle Source Reservoirs. in \u003cem\u003eNoble Gases in Geochemistry and Cosmochemistry: Reviews in Mineralogy and Geochemistry\u003c/em\u003e 247\u0026ndash;318 (2002).\u003c/li\u003e\n\u003cli\u003eGraham, D. W., Michael, P. J. \u0026amp; Rubin, K. H. An investigation of mid-ocean ridge degassing using He, CO2, and \u0026delta;13C variations during the 2005\u0026ndash;06 eruption at 9\u0026deg;50\u0026prime;N on the East Pacific Rise. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e504\u003c/strong\u003e, 84\u0026ndash;93 (2018).\u003c/li\u003e\n\u003cli\u003eBatiza, R., Niu, V. \u0026amp; Zayac, W. C. Chemistry of seamounts near the East Pacific Rise: Implications for the geometry of subaxial mantle flow. \u003cem\u003eGeology\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 1122\u0026ndash;1125 (1990).\u003c/li\u003e\n\u003cli\u003eFornari, D. J., Perfit, M. R., Allan, J. F. \u0026amp; Batiza, R. Small-scale heterogeneities in depleted mantle sources: near-ridge seamount lava geochemistry and implications for mid-ocean-ridge magmatic processes. \u003cem\u003eLetters to Nature\u003c/em\u003e \u003cstrong\u003e331\u003c/strong\u003e, 511\u0026ndash;513 (1988).\u003c/li\u003e\n\u003cli\u003eHall, L. S., Mahoney, J. J., Sinton, J. M. \u0026amp; Duncan, R. A. Spatial and temporal distribution of a C-like asthenospheric component in the Rano Rahi Seamount Field, East Pacific Rise, 15\u0026deg;-19\u0026deg;S. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1\u0026ndash;27 (2006).\u003c/li\u003e\n\u003cli\u003eShimizu, K. \u003cem\u003eet al.\u003c/em\u003e Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: Implications for the volatile content of the Pacific upper mantle. \u003cem\u003eGeochim Cosmochim Acta\u003c/em\u003e \u003cstrong\u003e176\u003c/strong\u003e, 44\u0026ndash;80 (2016).\u003c/li\u003e\n\u003cli\u003ePerfit, M. R. \u003cem\u003eet al.\u003c/em\u003e Small-scale spatial and temporal variations in mid-ocean ridge crest magmatic processes. \u003cem\u003eGeology\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 375\u0026ndash;379 (1994).\u003c/li\u003e\n\u003cli\u003eWaters, C. L., Sims, K. W. W., Perfit, M. R., Blichert-Toft, J. \u0026amp; Blusztajn, J. Perspective on the genesis of E-MORB from chemical and isotopic heterogeneity at 9-10\u0026deg;n East Pacific Rise. \u003cem\u003eJournal of Petrology\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 565\u0026ndash;602 (2011).\u003c/li\u003e\n\u003cli\u003eNiu, Y., Collerson, K. D., Batiza, R., Wendt, J. I. \u0026amp; Regelous, M. Origin of enriched-type mid-ocean ridge basalt at ridges far from mantle plumes: The East Pacific Rise at 11\u0026deg;20\u0026prime;N. \u003cem\u003eJ Geophys Res Solid Earth\u003c/em\u003e \u003cstrong\u003e104\u003c/strong\u003e, 7067\u0026ndash;7087 (1999).\u003c/li\u003e\n\u003cli\u003ePrinzhofer, A., Lewin, E. \u0026amp; Allegre, C. J. Stochastic melting of the marble cake mantle: evidence from local study of the East Pacific Rise at 12 50\u0026rsquo; N. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e92\u003c/strong\u003e, 189\u0026ndash;206 (1989).\u003c/li\u003e\n\u003cli\u003eHanyu, T., Kaneoka, I. \u0026amp; Nagao, K. Noble gas study of HIMU and EM ocean island basalts in the Polynesian region. \u003cem\u003eGeochim Cosmochim Acta\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 1181\u0026ndash;1201 (1999).\u003c/li\u003e\n\u003cli\u003eKimura, J. I., Gill, J. B., Skora, S., van Keken, P. E. \u0026amp; Kawabata, H. Origin of geochemical mantle components: Role of subduction filter. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 3289\u0026ndash;3325 (2016).\u003c/li\u003e\n\u003cli\u003eShimoda, G. \u0026amp; Kogiso, T. Effect of Serpentinite Dehydration in Subducting Slabs on Isotopic Diversity in Recycled Oceanic Crust and Its Role in Isotopic Heterogeneity of the Mantle. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 5449\u0026ndash;5472 (2019).\u003c/li\u003e\n\u003cli\u003eChauvel, C. \u003cem\u003eet al.\u003c/em\u003e Fani Maor\u0026eacute;, a new \u0026ldquo;young HIMU\u0026rdquo; volcano with extreme geochemistry. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e626\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eHoernle, K., Tilton, G., Le Bas, M. J., Duggen, S. \u0026amp; Garbe-Sch\u0026ouml;nberg, D. Geochemistry of oceanic carbonatites compared with continental carbonatites: Mantle recycling of oceanic crustal carbonate. \u003cem\u003eContributions to Mineralogy and Petrology\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 520\u0026ndash;542 (2002).\u003c/li\u003e\n\u003cli\u003ePlank, T. \u0026amp; Langmuir, C. H. The chemical composition of subducting sediment and its consequences for the crust and mantle. \u003cem\u003eChem Geol\u003c/em\u003e \u003cstrong\u003e145\u003c/strong\u003e, 325\u0026ndash;394 (1998).\u003c/li\u003e\n\u003cli\u003eGli\u0026scaron;ović, P. \u0026amp; Forte, A. M. A new back-and-forth iterative method for time-reversed convection modeling: Implications for the Cenozoic evolution of 3-D structure and dynamics of the mantle. \u003cem\u003eJ Geophys Res Solid Earth\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 4067\u0026ndash;4084 (2016).\u003c/li\u003e\n\u003cli\u003eGli\u0026scaron;ović, P. \u0026amp; Forte, A. M. Reconstructing the Cenozoic evolution of the mantle: Implications for mantle plume dynamics under the Pacific and Indian plates. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e390\u003c/strong\u003e, 146\u0026ndash;156 (2014).\u003c/li\u003e\n\u003cli\u003eGli\u0026scaron;ovic, P. \u0026amp; Forte, A. M. Two deep-mantle sources for Paleocene doming and volcanism in the North Atlantic. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 13227\u0026ndash;13232 (2019).\u003c/li\u003e\n\u003cli\u003eJohnston, G., Forte, A. \u0026amp; Glisovic, P. Time-Dependent Convective Modelling of DUPAL Signatures in the South Atlantic: Implications for LLSVP Source Region. in \u003cem\u003eAGU Fall Meeting\u003c/em\u003e DI44B-07 (New Orleans, Louisiana, 2021).\u003c/li\u003e\n\u003cli\u003eMitrovica, J. X. \u0026amp; Forte, A. M. A new inference of mantle viscosity based upon joint inversion of convection and glacial isostatic adjustment data. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e225\u003c/strong\u003e, 177\u0026ndash;189 (2004).\u003c/li\u003e\n\u003cli\u003eForte, A. M. \u003cem\u003eet al.\u003c/em\u003e Joint seismic-geodynamic-mineral physical modelling of African geodynamics: A reconciliation of deep-mantle convection with surface geophysical constraints. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e295\u003c/strong\u003e, 329\u0026ndash;341 (2010).\u003c/li\u003e\n\u003cli\u003eMann, P., Rogers, R. D. \u0026amp; Gahagan, L. Overview of plate tectonic history and its unresolved tectonic problems. in \u003cem\u003eCentral America: geology, resources, and hazards\u003c/em\u003e 205\u0026ndash;241 (2007).\u003c/li\u003e\n\u003cli\u003eGoes, S., Agrusta, R., van Hunen, J. \u0026amp; Garel, F. Subduction-transition zone interaction: A review. \u003cem\u003eGeosphere\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 644\u0026ndash;664 (2017).\u003c/li\u003e\n\u003cli\u003eCabral, R. A. \u003cem\u003eet al.\u003c/em\u003e Anomalous sulphur isotopes in plume lavas reveal deep mantle storage of Archaean crust. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e496\u003c/strong\u003e, 490\u0026ndash;493 (2013).\u003c/li\u003e\n\u003cli\u003eDottin, J. W., Labidi, J., Lekic, V., Jackson, M. G. \u0026amp; Farquhar, J. Sulfur isotope characterization of primordial and recycled sources feeding the Samoan mantle plume. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e534\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eFarquhar, J. \u0026amp; Jackson, M. Missing Archean sulfur returned from the mantle. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e113\u003c/strong\u003e, 12893\u0026ndash;12895 (2016).\u003c/li\u003e\n\u003cli\u003eKogiso, T., Tatsumi, Y., Shimoda, G. \u0026amp; Barsczus, H. G. High \u0026mu; (HIMU) ocean island basalts in southern Polynesia: New evidence for whole mantle scale recycling of subducted oceanic crust. \u003cem\u003eJ Geophys Res Solid Earth\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 8085\u0026ndash;8103 (1997).\u003c/li\u003e\n\u003cli\u003eWei, X. \u003cem\u003eet al.\u003c/em\u003e Co-Occurrence of HIMU and EM1 Components in a Single Magellan Seamount: Implications for the Formation of West Pacific Seamount Province. \u003cem\u003eJournal of Petrology\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 1\u0026ndash;19 (2022).\u003c/li\u003e\n\u003cli\u003eHauff, F., Hoernle, K. \u0026amp; Schmidt, A. Sr-Nd-Pb composition of Mesozoic Pacific oceanic crust (Site 1149 and 801, ODP Leg 185): Implications for alteration of ocean crust and the input into the Izu-Bonin-Mariana subduction system. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, (2003).\u003c/li\u003e\n\u003cli\u003eErhardt, A. M. \u003cem\u003eet al.\u003c/em\u003e Assessing Sedimentary Detrital Pb Isotopes as a Dust Tracer in the Pacific Ocean. \u003cem\u003ePaleoceanogr Paleoclimatol\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eHoernle, K. Geochemistry of Jurassic Oceanic Crust beneath Gran Canaria (Canary Islands): Implications for Crustal Recycling and Assimilation. \u003cem\u003eJOURNAL OF PETROLOGY\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 859\u0026ndash;880 (1998).\u003c/li\u003e\n\u003cli\u003eHanan, B. B. \u0026amp; Graham, D. W. Lead and Helium Isotope Evidence from Oceanic Basalts for a Common Deep Source of Mantle Plumes. \u003cem\u003eScience (1979)\u003c/em\u003e \u003cstrong\u003e272\u003c/strong\u003e, 991\u0026ndash;995 (1996).\u003c/li\u003e\n\u003cli\u003eSalters, V. J. M. \u0026amp; Stracke, A. Composition of the depleted mantle. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1\u0026ndash;27 (2004).\u003c/li\u003e\n\u003cli\u003eBach, W., Bernhard, P. E., Hart, S. R. \u0026amp; Blusztajn, J. S. Geochemistry of hydrothermally altered oceanic crust: DSDP/ODP Hole 504B-Implications for seawater-crust exchange budgets and Sr-and Pb-isotopic evolution of the mantle. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1\u0026ndash;29 (2003).\u003c/li\u003e\n\u003cli\u003eGraham, D. W. \u0026amp; Michael, P. J. Predominantly recycled carbon in Earth\u0026rsquo;s upper mantle revealed by He-CO2-Ba systematics in ultradepleted ocean ridge basalts. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e554\u003c/strong\u003e, 116646 (2021).\u003c/li\u003e\n\u003cli\u003eLeybourne, M. I. \u0026amp; Van Wagoner, N. A. Heck and Heckle seamounts, northeast Pacific Ocean: high extrusion rates of primitive and highly depleted mid-ocean ridge basalt on off- ridge seamounts. \u003cem\u003eJ Geophys Res\u003c/em\u003e \u003cstrong\u003e96\u003c/strong\u003e, (1991).\u003c/li\u003e\n\u003cli\u003eWhite, W. M. Trace Elements. in \u003cem\u003eGeochemistry\u003c/em\u003e 258\u0026ndash;312 (2013).\u003c/li\u003e\n\u003cli\u003eGeorgen, J. E., Kurz, M. D., Dick, H. J. B. \u0026amp; Lin, J. Low 3He/4He ratios in basalt glasses from the western Southwest Indian Ridge (10\u0026deg;-24\u0026deg;E). \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e206\u003c/strong\u003e, 509\u0026ndash;528 (2003).\u003c/li\u003e\n\u003cli\u003eHamelin, C. \u003cem\u003eet al.\u003c/em\u003e Geochemical portray of the Pacific Ridge: New isotopic data and statistical techniques. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e302\u003c/strong\u003e, 154\u0026ndash;162 (2011).\u003c/li\u003e\n\u003cli\u003eAnderson, M. K. \u003cem\u003eet al.\u003c/em\u003e Constraints on near-ridge magmatism using 40Ar/39Ar geochronology of enriched MORB from the 8\u0026deg;20\u0026rsquo; N seamount chain. \u003cem\u003eEarth Planet Sci Lett\u003c/em\u003e \u003cstrong\u003e618\u003c/strong\u003e, 118278 (2023).\u003c/li\u003e\n\u003cli\u003eRomano, V. \u003cem\u003eet al.\u003c/em\u003e The formation of the 8˚20\u0026rsquo; N seamount chain, east pacific rise. \u003cem\u003eMarine Geophysical Research\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eKamenov, G. D., Perfit, M. R., Mueller, P. A. \u0026amp; Jonasson, I. R. Controls on magmatism in an island arc environment: Study of lavas and sub-arc xenoliths from the Tabar-Lihir-Tanga-Feni island chain, Papua New Guinea. \u003cem\u003eContributions to Mineralogy and Petrology\u003c/em\u003e \u003cstrong\u003e155\u003c/strong\u003e, 635\u0026ndash;656 (2008).\u003c/li\u003e\n\u003cli\u003eGraham, D. W., Hanan, B. B., H\u0026eacute;mond, C., Blichert-Toft, J. \u0026amp; Albar\u0026egrave;de, F. Helium isotopic textures in Earth\u0026rsquo;s upper mantle. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 2048\u0026ndash;2074 (2014).\u003c/li\u003e\n\u003cli\u003eSimmons, N. A., Forte, A. M., Boschi, L. \u0026amp; Grand, S. P. GyPSuM: A joint tomographic model of mantle density and seismic wave speeds. \u003cem\u003eJ Geophys Res Solid Earth\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, (2010).\u003c/li\u003e\n\u003cli\u003eGli\u0026scaron;ović, P., Forte, A. M. \u0026amp; Moucha, R. Time-dependent convection models of mantle thermal structure constrained by seismic tomography and geodynamics: Implications for mantle plume dynamics and CMB heat flux. \u003cem\u003eGeophys J Int\u003c/em\u003e \u003cstrong\u003e190\u003c/strong\u003e, 785\u0026ndash;815 (2012).\u003c/li\u003e\n\u003cli\u003eGli\u0026scaron;ović, P. \u0026amp; Forte, A. M. Importance of initial buoyancy field on evolution of mantle thermal structure: Implications of surface boundary conditions. \u003cem\u003eGeoscience Frontiers\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 3\u0026ndash;22 (2015).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003e8\u0026deg;20\u0026rsquo; N Seamount Sample Collection and Preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe 8\u0026deg;20\u0026rsquo; N seamount chain is an ~170 km long chain of volcanic ridges and seamounts oriented perpendicular to the EPR and north of the Siqueiros fracture zone (Figure S1). Two research expeditions on RV Atlantis (AT37-05 and AT42-06) in 2016 and 2018 sampled, mapped, and collected gravity and magnetic data across the chain. Two of the most geochemically heterogeneous seamounts, Oscar and Coral, were selected for highest-density sampling in 2018. Off-axis eruptions up to 90 km from the ridge axis are attributed to melt channelization from zones of weakness in the lithosphere related to the tectonic evolution of the nearby Siqueiros transform.\u003c/p\u003e\n\u003cp\u003eThe 8\u0026deg;20\u0026rsquo; N seamounts form an E-W trending chain of volcanoes and constructional volcanic ridges extending ~170 km perpendicular from the northern East Pacific Rise. The chain is northwest of the Siqueiros transform and directly north of the Siqueiros fracture zone off-axis. Two research expeditions (AT37-05 and AT42-06) visited the 8\u0026deg;20\u0026rsquo; N seamounts in 2016 and 2018 respectively, to conduct high spatial-resolution sampling using HOV \u003cem\u003eAlvin\u003c/em\u003e (sample names starting with ALV) and dredges (sample names starting with OS). Although the 2016 cruise sampled the full length of the chain, the 2018 cruise focused on more detailed sampling along Oscar and Coral seamounts. Combined, the research expeditions collected 305 basalt samples with glassy rinds, all of which have been analyzed for major elements, more than half for trace elements (184 samples), 72 for Pb, Sr, and Nd isotopes, and 36 for He isotopes. The major and trace element concentrations and Pb, Sr, and Nd isotopes from the 2016 expedition are used to test models for melting heterogeneous mantle\u003csup\u003e13\u003c/sup\u003e. New \u003csup\u003e40\u003c/sup\u003eAr/\u003csup\u003e39\u003c/sup\u003eAr geochronological and geophysical observations (magnetic and gravity data) from the 8\u0026deg;20\u0026rsquo; N seamounts tie the seamount magmatism to the nearby Siqueiros and its complex tectonic evolution over the past ~3.8 myrs\u003csup\u003e85,86\u003c/sup\u003e. This paper is the first publication to include any He isotope data from the seamounts and new samples collected in 2018.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThe basalt samples are mostly clean aphyric basalts with glassy rinds useful for preserving the liquid composition of magma at the time of eruption. Alteration and phenocryst-free glasses were ultrasonically cleaned in deionized water.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrace Element Analytical Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTrace element concentrations were determined on 148 samples using a Laser Ablation (UP213 Nd-YAG New Wave Research laser) ThermoElectron X-Series II Quadrupole Inductively Coupled Plasma Mass Spectrometer (LA-ICP-MS) at Boise State University\u003csup\u003e13\u003c/sup\u003e. The remaining 35 samples were measured on solutions using an Element II High-Resolution Inductively Coupled Plasma Mass Spectrometer at University of Florida. About 50 mg of clean alteration-free glass for each sample was digested in sealed Teflon vials at 100\u0026deg; C in HF-HNO\u003csub\u003e3\u003c/sub\u003e mixture for 24 hours. The solutions were evaporate dried, then acidified with 6N HCl while sealed at 100\u0026deg; C overnight. After evaporate drying again, the samples were diluted with 4.5 mL of 8 ppb Re-Rh spiked 5% HNO\u003csub\u003e3\u0026nbsp;\u003c/sub\u003eand sealed at 100\u0026deg; C overnight. The samples were analyzed using Re and Rh as internal standards, and results were corrected using USGS standards AGV-1 and ENDV (resulting in a correction factor of 1.08 and 0.84 for Lu and Pb respectively). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePb, Nd, and Sr isotope Analytical Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRadiogenic isotopes were collected on 72 basalts from the 8\u0026deg;20\u0026rsquo; N seamounts at the University of Florida following previously established dissolution and chromatographic procedures\u003csup\u003e13,38\u003c/sup\u003e. Pb, Sr, and Nd isotopes were measured using a Nu-Plasma HR multicollector (MC) ICP-MS following methods described in refs\u003csup\u003e38,87\u003c/sup\u003e. Standards NBS-981, NBS-987, and JNdi-1 were run every 5-6 samples for Pb, Sr, and Nd respectively, and averages are reported in Table S3.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHelium Analytical Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHelium concentrations and \u003csup\u003e3\u003c/sup\u003eHe/\u003csup\u003e4\u003c/sup\u003eHe ratios in basalt glass were determined at Oregon State University using in-vacuo crushing experiments followed by noble gas mass spectrometry\u003csup\u003e46,88\u003c/sup\u003e. Several clean, alteration-free chips of glass were loaded into stainless steel crushers where they were crushed 75 times while connected to the vacuum line. Released CO\u003csub\u003e2\u003c/sub\u003e and H\u003csub\u003e2\u003c/sub\u003eO were condensed together in a U-trap held at 77 K (liquid N\u003csub\u003e2\u003c/sub\u003e temperature) and non-condensable reactive gases were removed using SAES Zr-Al getters. Noble gases were cryogenically separated using a charcoal trap at 10 K. The trap was heated to 45 K, retaining neon and heavier noble gases and releasing helium directly into the Nu Noblesse Mass Spectrometer for isotope ratio and peak height measurements. Line blanks performed prior to sample analyses were typically \u0026lt; 5x10\u003csup\u003e-11\u003c/sup\u003e cm\u003csup\u003e3\u003c/sup\u003e STP \u003csup\u003e4\u003c/sup\u003eHe.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSMB-GLOSS Isotopic Mixing Calculations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on trace element estimates for subduction-modified basalts (SMB\u003csup\u003e27\u003c/sup\u003e) and isotope ratios for altered oceanic crust (AOC\u003csup\u003e74\u003c/sup\u003e), and global subducting sediment trace elements and isotope ratios (GLOSS\u003csup\u003e60\u003c/sup\u003e), we approximate the full range of possible Pb, Sr, and Nd isotope ratios for a subducted mixture of 99% SMB + 1% GLOSS required for our isotopic evolution model. While it is challenging to perfectly reconstruct a mixture consistent across all isotopic systems, we demonstrate in Table S2 a range of scenarios capable of producing a protolith required to form the HIMU within 130 Ma. In general, assuming there is very little sediment addition (1-2%) to subducting oceanic crust, Pb would be most significantly leveraged by sediments, while Sr and Nd would be less so. In the simplest scenario, we assume 1% average GLOSS Pb, Sr, and Nd are mixed with 99% of SMB, which successfully produces \u003csup\u003e206\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb, \u003csup\u003e207\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb, and \u003csup\u003e208\u003c/sup\u003ePb/\u003csup\u003e204\u003c/sup\u003ePb consistent with ratios our reverse Pb model predicts subducted ~130 Ma. However, in this simplified scenario using the GLOSS average, \u003csup\u003e87\u003c/sup\u003eSr/\u003csup\u003e86\u003c/sup\u003eSr and \u003csup\u003e143\u003c/sup\u003eNd/\u003csup\u003e144\u003c/sup\u003eNd of this mixture are too low to be reasonable protoliths. However, as ref\u003csup\u003e60\u003c/sup\u003e demonstrated, there are extremely large ranges of Pb, Sr, and Nd concentrations and isotope ratios even between trenches around the globe. To form our HIMU Sr and Nd ratios in 130 Ma, we instead require subducting sediment compositions closer to the lower end of GLOSS (Table S2), such as pelagic sediments more consistent with a mid-ocean arc distal to a continental sediment source.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eParticle Tracking Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe reconstruct the past 3D structure of Earth\u0026rsquo;s mantle using numerical models of backward convective flow, with present-day mantle tomography as the initial condition\u003csup\u003e61,62\u003c/sup\u003e. By reversing time in the convection equations, we can trace the evolution of thermal anomalies and track how mantle parcels move from the present day back to earlier geologic epochs. This time-reversal approach employs convection models in which temperature and velocity fields are iteratively adjusted so that, when integrated forward again, they remain consistent with present-day tomography and (optionally) geologic constraints on plate motions. These methods provide a self-consistent framework that bridges global seismic tomography, mineral-physical constraints, geodynamic data, and plate reconstructions\u003csup\u003e63\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eModel Setup\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur primary input is a 3D temperature anomaly field derived from the GyPSuM global tomography model\u003csup\u003e89\u003c/sup\u003e. Because tomography constrains lateral variations but not the horizontally averaged temperature (geotherm), we superimpose these anomalies on an adiabatic background (1600 K at the surface to ~2456 K at the core\u0026ndash;mantle boundary: CMB). Key reference characteristics of the mantle (viscosity, density, thermal conductivity, etc.) are taken from refs\u003csup\u003e61,90\u003c/sup\u003e. Mantle internal heat sources (radioactive + secular cooling) sum to ~24 TW.\u003c/p\u003e\n\u003cp\u003eWe employ two depth-dependent viscosity profiles that are constrained by geodynamic observables. V1 model\u003csup\u003e65\u003c/sup\u003e has a lower viscosity and a thinner lithosphere, well-suited for modelling convective flow below oceanic plates and mid-ocean ridge regimes. V2 model\u003csup\u003e66\u003c/sup\u003e has an overall higher-viscosity mantle, thicker lithosphere, a low-viscosity layer at ~220 km, and a stiffer lower mantle. These viscosity models have been validated against mineral-physics constraints and surface geodynamic data\u003csup\u003e63,65,66\u003c/sup\u003e. Varying viscosity between V1 and V2 mainly shifts the timescales of mantle transport but does not substantially alter the large-scale geometry of the flow paths.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eForward and Backward Convection\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn forward mode, we implemented a pseudo-spectral solution of the conservation of mass, momentum, and energy for a Newtonian viscous fluid in a compressible, self-gravitating mantle\u003csup\u003e90\u003c/sup\u003e. Time-reversed flow calculations are numerically challenging due to the irreversibility of diffusion. We adopt the modified quasi-reversibility (QRV) technique\u003csup\u003e62\u003c/sup\u003e, which introduces a small biharmonic diffusion term to stabilize backward integration. We vary the regularization parameter with time, following the scheme in ref\u003csup\u003e61\u003c/sup\u003e, ensuring minimal unphysical diffusion while preventing exponential blow-up. To extend the accuracy of our time-reversed reconstructions, we couple the QRV method with the \u0026lsquo;Back-and-Forth Nudging\u0026rsquo; (BFN) technique\u003csup\u003e61\u003c/sup\u003e. This iterative method alternates between forward and backward integrations of the mantle convection model, spanning 2.5 Myr time windows, and iteratively updates the temperature field at each time step to minimize cumulative errors associated with thermal diffusion. The BFN method preserves the resolution of mantle heterogeneity given by the present-day tomography model, and it allows for more precise reconstructions of mantle dynamics extending into the geologic past, often to 70 Ma and as far back as 250 Ma (see supplementary discussion A-C).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eBoundary Conditions, Model Resolution, and Plate Reconstructions\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA plate-like mechanical boundary is applied at Earth\u0026rsquo;s surface, so the mantle\u0026rsquo;s buoyancy forces drive the plates rather than the plates imposing motion on the underlying mantle. Plate geometries over the Cenozoic are based on 5 Myr interval reconstructions in the Indo-Atlantic hotspot reference frame\u003csup\u003e29,30\u003c/sup\u003e. The models use: spherical harmonic expansions to degree 170 for horizontal resolution, yielding a horizontal resolution of ~64 km at the CMB and ~117 km near the surface; Chebyshev expansions up to order 129 radially, yielding a vertical resolution of 0.4 km near boundary layers to ~35 km in mid-mantle).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eParticle Tracking\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOnce the flow fields (\u0026ldquo;mantle wind\u0026rdquo;) are established, we compute particle flow lines via an eighth-order Dormand\u0026ndash;Prince Runge\u0026ndash;Kutta scheme that was implemented in ref\u003csup\u003e63\u003c/sup\u003e. These particle tracking calculations benefit from the enhanced precision afforded by the spectral method, which yields accurate calculations of the flow velocity at any point in the mantle domain. We note that deep mantle flow is relatively stable, preserving broad features (notably the \u0026ldquo;superplume\u0026rdquo; upwellings) for tens to hundreds of millions of years\u003csup\u003e90,91\u003c/sup\u003e. For times before 70 Ma, we often approximate deep flow as quasi-steady because the high viscosity in the lower mantle stabilizes long-wavelength downwelling and upwelling structures\u003csup\u003e\u0026nbsp;90,91\u003c/sup\u003e. We compare these results to time-reversed mantle reconstructions that extend back to 250 Ma, which are subject to significant modelling uncertainties. A discussion of these associated uncertainties is presented in the supplementary discussion A.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eApplication to the East Pacific Rise\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe applied this approach to a mantle parcel at 8.409454 N, \u0026minus;104.412 E, 50 km beneath the 8\u0026deg;20\u0026rsquo; N seamount chain near the East Pacific Rise (EPR). Under the V1 (low viscosity) profile, this particle descends to ~2500 km depth (~387 km above the CMB) in ~75 Myr (Figure S3). It then returns to the upper mantle at ~130 Ma, near \u0026minus;96\u0026deg; E, 15\u0026deg; N. These tracks align with independent reconstructions showing the paleo-Mezcalera plate subducting beneath the paleo-Farallon plate\u003csup\u003e67\u003c/sup\u003e, forming the Caribbean\u0026ndash;Guerrero arc. Using the V2 (higher viscosity) profile almost doubles the transport time (~240 Ma to reach the same depth). However, both profiles converge on the same subduction-zone source region, demonstrating that while viscosity changes the timescale, it does not alter the essential pathway. Prior time-dependent studies\u003csup\u003e29,62\u003c/sup\u003e similarly suggest a long-lived upwelling beneath the southern half of the EPR, consistent with our backward tracking results. Hence, two scenarios could explain the recycled material feeding the EPR: (1) Rapid dredging of ancient deep-lower-mantle heterogeneities (~387 km above the CMB); (2) Oceanic crust subducted during the Early Cretaceous that transited the entire mantle over ~130 Myr. In either case, these particle tracks impose new lower-limit constraints on elemental cycling in Earth\u0026rsquo;s mantle. Combined with radiogenic isotope and trace element modeling, they support a role for subduction-driven heterogeneities in generating the notable geochemical signals at 8\u0026deg;20\u0026rsquo; N seamounts.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Mantle heterogeneity, East Pacific Rise, off-axis seamounts, seismic tomography, mantle convection, particle tracking, radiogenic isotopes, trace elements","lastPublishedDoi":"10.21203/rs.3.rs-6173319/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6173319/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCompositionally diverse lavas from the 8\u0026deg;20\u0026rsquo; N seamount chain near the East Pacific Rise (EPR) record greater geochemical heterogeneity than observed along the entire northern EPR, including evidence for a 4-component mantle, and the first discovery of high-\u0026micro; (high time-integrated \u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e204\u003c/sup\u003ePb) mantle source at the northern EPR. The amplitude of Pb, Sr, and Nd isotopic heterogeneity is on the scale of the Galapagos plume but in an area distal to a mantle plume. Our study combines isotopic and trace element compositions of near-EPR seamounts with recent mantle particle tracking models to identify deeply subducted origins for northern EPR mantle heterogeneity. We show that geochemical models (using He, Pb, Sr, Nd isotopes and trace elements) and geodynamic models (using time-dependent mantle flow simulations based on tomography) are self-consistent with tectonically recycled material feeding the northern EPR asthenosphere. These models provide a rare opportunity to assess geographic origins of mantle heterogeneity, tying the chemical and isotopic variability at the northern EPR to a paleo subduction zone at ~\u0026thinsp;130 Ma, and reveal that one of the fastest spreading ridges on Earth samples deeply recycled components associated with whole mantle convection.\u003c/p\u003e","manuscriptTitle":"Deeply Recycled Origins of Mantle Heterogeneity at the Northern East Pacific Rise","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-20 07:30:15","doi":"10.21203/rs.3.rs-6173319/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"c5108073-ba66-43bc-a8a6-44a13ba27e81","owner":[],"postedDate":"June 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":49538538,"name":"Earth and environmental sciences/Solid Earth sciences/Geochemistry"},{"id":49538539,"name":"Earth and environmental sciences/Solid Earth sciences/Geodynamics"},{"id":49538540,"name":"Earth and environmental sciences/Solid Earth sciences/Petrology"}],"tags":[],"updatedAt":"2025-06-20T07:30:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-20 07:30:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6173319","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6173319","identity":"rs-6173319","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}